WHAT MAKES A SYSTEM COMPLEX? AN APPROACH TO SELF ORGANIZATION AND EMERGENCE

Michel Cotsaftis LACSC/ECE [email protected]

“Men in their arrogance claim to understand the nature of Creation, and devise elaborate theories to describe its behaviour. But always they discover in the end that God is more clever than they thought Sister Miriam Godwinson

The fast changing reality in technical and natural domains perceived by always more accurate observations has drawn the attention on a new and very broad class of systems mainly characterized by specific behaviour which has been entered under the common wording “complexity”. Based on elementary system graph representation with components as nodes and interactions as vertices, it is shown that systems belong to only three states : simple, complicated, and complex, the main properties of which are discussed. The first two states have been studied at length over past centuries, and the last one finds its origin in the elementary fact that when system performance is pushed up, there exists a threshold above which interaction between components overtake outside interaction. At the same time, system self-organizes and filters corresponding outer action, making it more robust to outer effect, with emergence of a new behaviour which was not predictable from only components study. Examples in Physics and Biology are given, and three main classes of “complexity” behaviour are distinguished corresponding to different levels of difficulty to handle the problem of their dynamics. The great interest of using complex state properties in man-made systems is stressed and important issues are discussed. They mainly concentrate on the difficult balance to be established between the relative system isolation when becoming complex and the delegation of corresponding new capability from (outside) operator. This implies giving the system some “intelligence” in an adequate frame between the new augmented system state and supervising operator, with consequences on the canonical system triplet {effector-sensor-controller} which has to be reorganized in this new setting. Moreover, it is observed that entering complexity state opens the possibility for the function to feedback onto the structure, ie to mimic at technical level the invention of Nature over Her very long history.

1. Introduction

After a very long period of observations, it became very slowly clear to first Human observers since prehistoric age that the phenomena they were distinguishing were due to a specific order in Nature and not to the capricious will of mysterious Gods. Since then, they have been over the centuries patiently organizing the most visible ones, and, proposing hypotheses to find this order, they finally end up on laws to represent the simplest phenomena first which have been verified in the framework of actual observations. These elementary phenomena were mainly concerning “simple” systems which can reasonably be isolated in their dynamics and their observation. In the mean time, the advance in technology and observation accuracy was driving the attention on more complicated systems with always larger number of elements, which in the 17th century have been shown by R. Descartes to be reducible in their study to the former class of simple ones[1]. Finally, the result today of this long quest was an “adapted” representation of the Universe with a broad “classical” part for “human size” phenomena, corrected by “quantum” effects at very small atomic level, and by “relativistic” ones at very large galactic level. This basic “1-dimensional” picture emerging from millennium long effort of human kind, and leaving fundamental questions only at infinitely large universe and infinitely small super elementary particle levels, is however not sufficient to represent correctly enough the present situation where systems exist over an extremely large parametric domain now easily explored with the recent development of always more performing technology over the last decades. First observed from time to time in different domains, there is now evidence of existence of new huge class of systems, natural as well as artificial in all scientific and technical human activities, which have reached their own status by the corner of the millennium under the name of “complex” systems. They are in some sense opening a “2-dimensional” more global picture of the universe by being located along a direction “perpendicular” to previous 1-dimension line going from infinitely small to infinitely large size, see Fig.1. There is today a strong questioning about their origin and their formation[2]. Some have even suggested that previous “mechanistic” approach to the universe by researching “ultimate” fully unifying law[3] is just mistaking their very and irreducible structure and that a new “revolution” is necessary to grasp all aspects of known universe today[4]. To address these questions, a very pedestrian approach is proposed here, based on elementary source-sink model applied to the graph representing the aggregate of possible components of a system. Then it evidently appears that system structure can be divided into three different groups, simple, complicated and complex, corresponding to specific properties. The first two groups are usual ones nicely approachable by the methods of scientific reductionism[5]. The third group, by its very global nature, cannot be just reduced to the effect of its components[6] and requires some adjustment for being correctly handled, because now the key point is the way the system behaves under (or against) the action of its environment. It will appear that the mechanistic notion of individual “trajectory” is loosing its meaning and should be replaced by more general “manifold” entity corresponding to accessible “invariants” under environment action. System self-organization and emergence properties are deduced and discussed. Different grades of complexity are fixed depending on specific system properties which relate to well known classes of observed phenomena. Advantages for application of complex structure to man made systems are stressed.

2. System State Analysis

Let consider first a system with a finite number N of identified and distinguishable components which can be represented by a graph with N nodes Ni . On this graph, three types of vertices can be drawn in between the N components i and outside sources e whenever an exchange exists between them. They correspond respectively to free flight state for vertex Vii, to driven state from outer source for vertex Vie, and to interactive state with other system components for vertex Vij., see Fig.2. System dynamics are usually resulting from a combination of previous three different exchanges to which, on a very general setting, there can be associated three characteristic fluxes for each system component. Their specific nature (power, information, chemical,..) is depending on the problem at hand, but is here unnecessary as long as they unambiguously characterize system components status. The first flux corresponds to the ‘’free’’ dynamics of i-th component pii along vertex Vii, the second one pi,e to the transfer flux between outer source and system i-th component along vertex Vie, and the last one to inter components effects pi,int = Infj{pi,j}, where pi,j is the characteristic and oriented flux exchange between components i and j along vertex Vij, see Fig.2. Depending on the various possible ordering between these three fluxes, the system behaves in different ways. When (A) pii >> pi,int, pi,e the i-th component is weakly coupled to other components. Its dynamics are mainly the ones of an isolated element slightly perturbed by the other actions, and the system reduces to a set of almost independent one- component sub-systems. In this sense it can be considered as a simple one. When (B) pi,e >> pi,int, pii the i-th component is mainly depending on outside sources. The action of the other components creates a (weak) coupling between the components, but the system can nevertheless be decoupled, at least locally, into a set of sub-systems which can be acted upon and controlled by as many exterior sources as there are components in the system because they can still be identified. In this sense the system can be termed as complicated from etymologic sense (from Latin cum pliare : piled up with). Finally when (C) pi,int >> pie, pii the i-th component is very strongly coupled to other ones and its dynamics are now mainly determined by components interaction satisfying the inequality. If apparently there does not seem to exist a dramatic difference with the previous case as concerns their dynamics, there is however a fundamental one as concerns the effect of outside action. In previous case input from the source can be tracked to the concerned component so this degree of freedom can be completely controlled from outside, whereas in present one the nature of interactions dominates and shields this tracking. As a result, control action can only be a ‘’global’’ one from other system components satisfying condition for second case (B) and system dynamics are now also driven by internal action. Consequently a totally new situation occurs for all system components passing in third state in the sense that their control cannot as in second case be fixed by only outer source action. Effectively, it is no longer possible as in second state to determine completely all system components dynamics from outside because of stronger interaction effect which now dominates the dynamics of concerned components. In a sense a self-organization has been taking place inside the system which leads to an internal control replacing the classical one from outside. So it is no longer possible to continue to manipulate inputs with as many degrees of freedom as the system initially has because of the mismatch which would result from the conflict with the internal control due to components interaction satisfying condition for the third case. External system control dimension is thus reduced. A system in this case will be termed as complex in agreement with etymology (from Latin cum plexus : tied up with). According to this definition, a very elementary test for determining if a system is passing to complex state is thus to verify that its control requires the manipulation of less degrees of freedom than the system has initially. There exists an immense literature about complexity, its definition and its properties covering an extremely wide range of domains from Philosophy to Technology[7], especially in recent years where its role has been “discovered” in many different fields such as networks now playing a crucial role with the ascent of Information Technologies[8]. Here emphasis is more modestly put only on more restricted complex state compared to complicated state as concerns action from outside environment (ie from control point of view) onto the system. For a system S with a finite number N of identified components, one can then define its index of complexity CS = 0 if it is only in complicated state, and CS = n/N if n < N components cluster in complex state by satisfying inequality (C) and become insensitive to outer action. CS measures the grasp on the system from outside action/control and the limit CS = 1 corresponds to totally autarchic system, the most complex possible structure with this number of components. This may look paradoxical as the definition of a simple system is precisely that it can be isolated. In fact this apparently contradictory statement is resulting from the very nature of internal interactions effect which reduces the number of invariants on which system trajectory takes place. An elementary example is given by the particles of a neutral gas for which their initial 6N positions and velocities (the mechanical invariants of motion) are reduced to the only energy (or temperature), justifying thermodynamic representation. In other words, when represented from outside, a system is the less complicated as it is more complex inside. Differently said, a complicated system remains complicated whether observed from inside or from outside, whereas when becoming complex it is less complicated when observed from outside, on top of the fact that being depending on a restricted number of parameters, it is more insensitive to outer action. However, when observing from outside a system with apparent CS = n/N, if N is not fixed when the system is not well known, there is no one-to-one relation in general between apparent outer behaviour of the system and its initial internal structure. It is only possible to check relative variation of CS when crossing condition (C) for some components. On a general setting, it is very easy from its definition to envision complex state as a powerful way set up by Nature to generate more sophisticated entities able to gain at the same time more independence with respect to their environment, or, in other terms, to become more robust to its variations and yet, to appear much less complicated outside, so that they can in turn link more easily to similar entities and generate a chain of nested clusters. This seems to be an underlying trend in evolution on Earth which has been allowing the elaboration of living beings, the up-most advanced form of complex systems[9]. Summarizing, exactly like matter does appear in three states : solids, liquids and gas, depending on conditions, systems are exhibiting three states : simple, complicated and complex. A very simple way to locate a system with respect to them is to plot for each component i the three values {pii , pi,e , pi,int} in a three dimensional space where each direction represents a single state. Each component is then represented by a point Ci and the system by a cluster of N points, see Fig.3. Depending on where they are located in this space, it is evident to figure out the status of the system by inequalities (A),(B) and (C). Moreover, it is easy to check what will be the consequence of parameter variation, especially when crossing inequality (C), a point extremely important for control of man-made systems as seen later. In this picture, components for which inequality pi,int >> pie, pii holds are in fact an internal sub- cluster which cannot be further split from outside observation and action. As indicated above, the first two states have been observed and studied along past centuries. The fact is that with modern and very detailed observation diagnosis, complex state has been now very often observed in natural systems, and becomes the most common one in a broad range of phenomena. For artificial man-made systems, even if it were sometimes crossed in the past, it is only with recent progress of advanced technology that, when pushing systems performance for higher efficiency, the threshold for overtaking of internal interactions effect corresponding to third case is also very routinely over passed now. So complex behaviour is seen in many situations, and becomes an object of study in itself, with considerable consequences in technical and industrial applications it implies in conjunction with the corresponding demand of more delegation of decisional nature to the systems for better efficiency. From previous and relatively rough source-sink representation always applicable in a first instance, it is already possible to stress a deep difference existing between the first two states and last complex one. In first two states, the possibility exists to split the system into as many independent one- component systems in a first approximation, whereas this is impossible in third state where all interacting components have to be taken as a whole. In mathematical terms the consequence is that usual approximation methods developed for the first two states do not straightforwardly apply and have to be revised in order to handle the global aspect of the coordinated response of components in complex state. This difficulty is at the origin of today important computerized research undertaken on the problem.

3. Emergence

When going to more specific situations, other elements are also entering the description and it is interesting from elementary global description above to recover various situations which have been observed and analytically studied for different parameters value. First it is likely that in general the system is not always in a ‘’pure’’ state and often exhibits a mixed structure where some components are in one state and others in another one. An important example is the case of natural inhomogeneous but continuous natural systems such as fluids with non zero gradients in a domain. In these systems, fluctuations are universally observed the source of which is the free energy available in between this stationary equilibrium and complete (homogeneous) thermodynamic one. In present case, the free energy is coming from the space gradients related to medium non homogeneity. They have in general a very large range of fluctuations (roughly because the system has a very large number of components) which can be split into two groups depending on their wavelength compared to system characteristic gradient length. Typically, the fluctuations with large wavelength excited in the medium are in complicated state and, because they are sensitive to boundary conditions, can be observed and possibly acted upon as such from outside. Fluctuations with small wavelength on the other hand are generally in complex state due to parameter values. So under their strong interactions, and because they are much less sensitive to boundary conditions, they are loosing their phase and globally excite a leak out of the fluid (usually called a transport) manifested by an out-flux expressing the non equilibrium situation of the system, and which counteracts the input flux responsible of medium non homogeneity. The determination of these transports is a very important element for qualifying the behaviour of the system and is an active research problem studied worldwide. This feature is observed for all natural systems when they enter the dissipative branch. A natural system is called dissipative[10] when it is exchanging (particles and energy) fluxes with outside environment. Evidently the channels by which internal energy sources are related to these fluxes are playing a privileged role because they regulate the energy ultimately available for the system and finally determine its self-organized state[11]. Dissipative systems more generally only exist to the expense of these fluxes, and they evolve with parameter change − such as the power input − along a set of neighbouring states determined by branching due to bifurcations where internal structure changes in compatibility with boundary conditions and by following the principle of largest stability. So the picture of such systems is a transport system governing flux exchanges guided by the bifurcation system which, as a pilot, fixes the structure along which these exchanges are taking place. Finding the branching pattern thus entirely defines the possible states of the system and determines the fluctuation spectrum. Branching is found as nontrivial solutions of variation equations deduced from general system dynamics equations. Despite the physical origin of phenomena is here well identified, the analysis is still in progress in a very large number of situations, such as in thermonuclear plasma to determine correct parameter sizing for realizing “burning” conditions[12], and in neutral fluids where turbulence is not fully explained yet[13], whereas the problem has been cleared up for fluctuations in deformable solid bodies[14]. In most cases the difficulty stems from finding an adequate representation of the global effect of small wavelength fluctuations, especially when due to parameter values, they feedback on longer time scale large wavelength oscillations and modify their dynamics. In this case, there is a significant change of initial system dynamics due to passage in complex state of an internal part of the system generated by branching phenomenon. Moreover, it is very likely that modification of system dynamics is the more important as the non homogeneity zone is the less ‘’transparent’’ between the two domains above and below along power flow. This has been extended to the extreme with living cell systems which have been able to completely encapsulate within a filtering membrane (ie a steep gradient) a space domain where very specific “memory” DNA molecules are fixing the dynamics of inside system they control, with corresponding exchange across the membrane. It is easy to understand that conjunction of a gate and of adequate information is the most efficient way to deeply modify locally the effect of regular physics laws, as exemplified by the counter streaming motion of a piece of wood in a river with one stage locks when regulating their opening and closing. However, dissipative behaviour does not exhaust all possible situations and many other ones do not follow this pattern. Natural systems so far are exhibiting components with relatively elementary features (charge, mass, geometrical structure, chemical activity, wavelength, frequency..), but there are also cases where complex state occurs in systems with more sophisticated components, usually called “agents”. Examples are herds of animals, insect colonies, living cell behaviour in organs and organisms, and population activity in an economy. In all cases, when observed from outside the systems are exhibiting relatively well defined behaviours but a very important element missing in previous analysis is the influence of the goal the systems are seemingly aiming at. Very often the components of these systems are searching through a collective action the satisfaction of properties they cannot reach alone, and to represent this situation the specific word “emergence” has been coined[15]. The point is that it is now possible to return back to previous case and in a unified picture to envision the laws of Physics themselves as emerging phenomena. For instance for an ensemble of neutral particles with hard ball interactions, and beyond the threshold of rarefied gas, (ie when the Knudsen number Kn = λ/L is decreasing to 0 from the value 1, with λ the particle mean free path and L a characteristic length scale), the particle system is suddenly passing from a complicated to a complex state (due to the huge value of Avogadro number). Consecutive to overtaking of collective interaction by collisions (expressed by decrease of Knudsen number), it could be said that there is emergence of a pressure and a temperature, which, from a point of view outside the gas, summarizes perfectly well the representative parameters (the invariants) describing it at this global level and amply justifies the thermodynamic representation in this case, see next §. The elementary reason is that here the fast particle dynamics described by Boltzmann equation relaxes within extremely short collision time to its equilibrium Maxwell-Boltzmann-Gibbs distribution only depending on temperature (and the potential of applied forces if any), which justifies usual stochastic methods[16]. Of course it is tempting at this point to “forget” the underlying mechanical base and to decide that perfect gas law, if any, is the emergence directly resulting from billiard ball “properties” of interacting (agents) molecules. The difficulty in this case is that application of “ergodic” hypothesis[17], even if it can mathematically be invented from scratch, rests here heavily upon thorough analysis of molecule ensemble dynamics, ie on mechanical laws. Similar problem does exist at atom level itself where, after baryons are assembled from primitive quark particles below some threshold energy, protons and neutrons assemble in turn themselves below another lower energy threshold into ions with only mass and charge parameters, able to combine finally with electrons to create atoms. Speaking of “emergence of atoms” from underlying components does not add any more information to classical approach. So in an opposite direction, a more conservative and probably more appropriate way to account for the new situation is to just state that when becoming complex, systems are exhibiting emergence of self-organization out of which it has been verified that their new behaviour in complex state is always a consequence, without going into the details of this behaviour which is depending on specific system parameters. In fact it is quite elementary from previous source-sink model to understand that a key point is in the accuracy of modelling the components in complex state, as long as the resulting “invariants” which will grasp all system information for interaction with environment are directly depending on this modelling. This has been at the origin of a computer “blind” search where the agents are given properties and “emerging” behaviour is obtained in a bottom-up approach, sometimes in surprising compatibility with experimental observations[18], in parallel to theoretical analysis[19]. Of course following classical science reductionism, there remains to justify the choice of agent properties, and to show that they are in one-to-one relation with observed behaviour. As this does not seem always possible, a different holistic approach[20] has been proposed where the research of ultimate underlying elements from which everything is constructed by as many layered shells on top of one another, is considered as unnecessary, to the price of loosing prediction power. Finally an extremely important remark is that the logical chain :

{stimuli/parameter change} → {higher interactions between some system components} → {passage to complex state} → {system self organization} → {emergence of new behaviour} discussed here is nothing but the sequence leading to the final step of system evolution toward more independence, and which is the feedback of “function” onto “structure”, a specific property of living organisms explaining their remarkable survival capability by structure modification.

4. Mathematical Analysis of Complex Systems

The few previous examples from common sense observation illustrate the elements which have been described above and which are providing a general base for complex system paradigm. From atomic nucleus to galaxy natural systems are seen to be constituted by aggregates of identifiable components (which, as already stressed, can be themselves, at each observation level, aggregates of smaller components) with well defined properties. These aggregates have been said to exhibit a complex behaviour when interaction between the components −or some of them− is overtaking their interactions with exterior environment. Similarly living beings are exhibiting the same behaviour, as observed with gregarious species, and in artificial man-made systems the same phenomenon is occurring when the overtaking conditions are satisfied. This is for instance the case for high enough performance level systems because the components are then tightly packed, as for high torque compact electrical actuators. Despite an extremely large variety of possible situations there are few base interactive processes leading to complex self-organization. Three main types will be discussed: the reducible case, the instantaneous differential case, and the integro-differential case with past neighbourhood memory.

4A − Reducible case :

It is first intuitively expectable that the approach to full problem gets easier when time-space scales are very different for components going to complex state as compared to the ones of other system components. This is mostly the situation in dissipative systems where characteristic frequency and wavelength of the spectrum of complex system part are respectively much larger and much smaller than the ones of the other (non complex) part of the system. The main reason is as explained above that components in complex state are produced in the initial system by the effect of bifurcation determined from variation equations of the system and are corresponding to high mode numbers (and have small enough wavelength). For such modes, their saturation amplitude is fixed by their nonlinear interaction which overtakes the level they would have reached under the only effect of the source and, as well known, exhibit globally a fluctuation spectrum usually very difficult to determine analytically.

Let τs and τcp be respectively the characteristic response time of the non complex system part and of n complex modes part with τcp = ε τs. The simplest situation n > 1 corresponds to the case where the interactions are strong and frequent enough to kill corresponding modes dynamics. In this case and because there are usually many modes with small amplitude, complex system part dynamics can be reduced to ‘’thermodynamic’’ approximation with a random distribution of fluctuation spectrum. Then the initial system of equations describing the full system splits into a restricted set of identified modes perturbed by a random term. This unifying and powerful model is quite successful in science and engineering[21] but it does not cover all situations. A more difficult and yet common case is occurring when n = 1, for which small O(ε) modes dynamics have long enough coherence characteristic time comparable to τs. To account for this coherence, asymptotic expansion of small modes dynamics from complex part dynamic equations is required up to first order to be injected in non complex system part dynamics, because ε order term acting coherently over a time ε−1 produces a non negligible finite order term. The system again reduces to its non complex part but now modified by coherent effects of complex system part dynamics, see Fig.4. The unsolved situations in [12,13] are belonging to this case. However it will nevertheless be termed as reducible because full system dynamics can be projected onto non complex system subpart. It clearly illustrates the fact that, given a system with a fixed number of degrees of freedom and associated initial and boundary conditions, this number should increase at bifurcation crossing so other conditions would be required to determine the solution. In fact this is not the case here as new complex modes dynamics can be solved by asymptotic procedure and re-injected into non complex system part which obviously keeps the same number of degrees of freedom. The only constraint is on initial complex modes amplitude which, due to their smallness, obey central limit theorem and are randomly distributed. So the difference with previous case n > 1 is that only initial amplitudes values are random now and follow afterward their own dynamics whereas they were previously random for all time. In both cases the number of final degrees of freedom remains the same, so despite their number has apparently increased at bifurcation, in reality the system is still of the same dimension. The extremely important consequence is that for controlled man-made systems, when their dynamics cross a bifurcation point, the same number of controllers is still required but they should be adapted to explore a new larger function space. The main effect of the bifurcation has been to augment system response to a wider function space, and the motion takes place on a manifold of this new space so that at the end the effective number of degrees of freedom is remaining the same. As stressed earlier, this behaviour is the manifestation of system self− organization. The double mechanical-thermodynamic feature is extremely general and characterizes complex reducible systems where complex part is added by bifurcation crossing, see Fig.5. Again control of such systems (whether natural or man-made) is not classical because for n ≥ 1 the new modes created after bifurcation crossing, though sometimes observable, are not accessible from outside. Because of their definite effect on system dynamics, a new more balanced approach respecting internal system action has to be worked out, which, very generally, is aimed at replacing inability of control action by insensitivity to variation for inaccessible complex modes. Though apparently loosing some hand on such systems, it has been surprisingly possible along this line to find explicit conditions in terms of system parameters expressing somewhat contradictory high preciseness (by asymptotic stability condition) and strong robustness (against unknown system and environment parts)[22]. In this way system dynamics are finally controlled and asymptotic stability can be demonstrated, but in general the price to pay is a not necessarily decreasing exponential asymptotic type.

4B − Instantaneous Differential Case :

The reducible case mainly deals with systems in partly or totally complex state, as with bifurcation crossing by some varying parameter, and with small parameter ordering between time and space scales, so it is possible to proceed to asymptotic expansion. More generally, a system may be in complete complex state, examples of which are atomic nucleus, herd of animals, and galaxies. Despite their very different space sizes, the systems exhibit always the same base characteristic feature to finally depend on an extremely restricted number of parameters as compared to the aggregate of their initial components. Searching the way to extract directly the remaining ‘’control’’ parameters of such systems from their dynamics is a fundamental issue which today motivates a huge research effort worldwide, especially in relation with information networking. Extensive analytical and numerical study has been developed for differential systems of generic form

dX =++A(,tXvX , )λμ . FtXu (, , ) . StX (, ) (1) dt where X = col(X1, X2,.. Xn) is system state space, λ,μ are n-vector coupling parameters, and A(.,.,.), F(.,.,.), S(.,.) are three specific q1×q2−matrix terms (q1,q2 ≤ n) corresponding to isolated free flight, nonlinear internal interactions and source terms respectively (the linear and source terms in the right hand side of eqn(1) are here split apart to indicate their respective role). The other variables v=v(X,t) and u(X,t) in A and F function account more generally for the possibility to feedback evolution of X(t) onto their own dynamics as it often occurs in systems when splitting parameters into given and manipulated control ones. For fixed u(.,.), v(.,.), depending on the values of λ,μ components the system will be in simple, complicated or complex state described in §1. When increasing the components of λ the system runs into complex state, and it has been repeatedly observed, especially on systems close to Hamiltonian ones[23], that system representative point in n−dimensional state space follows a more and more chaotic trajectory when crossing bifurcation values and at the end fills up a complete domain[24]. Of course sensitivity is largest when the system exhibits resonances, ie is close to conservative, and adapted mathematical expansion methods have to be worked out[25]. Because systems are basically non integrable[26], this is a direct evidence of increasing effect of internal interactions which reduce system dynamics to stay on attractor manifold of degree p < n, so that system dynamics are now layered on this manifold. This also expresses the fact that trajectories on the remaining n−p dimension space are becoming totally indistinguishable (from outside) when taken care of by internal interactions of n−p components going to complex state. So system trajectories reorganize here in equivalence classes which cannot be further split, a dual way to express the fact that there exists an invariant manifold on which system trajectories are lying. It is easy to understand that continuing to control these components by regular previous control[27] worked out for complicated state and specially designed for tracking a prescribed trajectory, is no longer possible and that a new approach is required which carefully respects internal system action due to complex state self- organization. More global methods of functional analysis[28] related to function space embedding in adapted function spaces[29] by fixed point theorem[30] are now in order as shown for reducible case[31], because they are providing the correct framework to grasp the new structure of system trajectory which cannot be fully tracked as before. Basically the method is again to counteract impreciseness in an element by robustness to its variation, a method very largely followed by living organisms.

4C − Integro-Differential Case with Past Neighbouring Memory :

Another important dimension relates to the properties system components are given, and a very influential one is the range of inter-component interaction, because this determines completely the build up of system clustering when becoming complex. Obviously long range interactions are leading to more intricate response with more difficult analysis. Examples are stars in a galaxy, electromagnetic interactions between ions and electrons in a plasma, animals in a herd and social behaviour of human population in economic trading such as stock market with internet link. In all cases a new element is coming from the size of the neighbouring domain each system component is sensitive to, and implies a time extension to past neighbouring components trajectories, see Fig.6. So the resulting complex behaviour is more generally determined by interactive component effects over a past time interval and weighted according to their importance. In this case systems are obeying for t ≥ t0>0 delayed equations of the form

dX = FtXt(, (),(),, It ud ) (2) dt

dI ⎛⎞t =−∫∫ρ(λϕ ,tGtXt )⎜⎟ , ( ), ( t ') Xt ( ht ( ', Xt ( ')) dt ', λλ d (3) dt ⎝⎠−∞ with X = col(X1, X2,.. Xn), I = col(I1, I2,.. Ip), which clearly indicates the role of the past through intermediate “moments” I(t) gathering a set of elements G with a weight ρ(λ,t), each G being itself constructed on the past history of system trajectories X(t) with fading weight ϕ(t), and where F(.,.,.) : + + 0 − R1 ×Rn×Rp×U×D → Rn and G(.,.,.,.) : R1 ×Rn×C (R1 ,Rn)×Λ → Rp are specific vectors. Their expression generally include control parameters u(.)∈U which have to be stated explicitly by control law to fix the structure of eqns(2,3) and uncontrolled disturbing effects often represented by a noise d(.)∈D. Here the delay h(.,X(.))< t can be state-depending as it appears in the analysis of natural (biological) systems where information from the past can be shielded by the state itself. The mixing from the breath of trajectory coverage and from past memory basically creates a “double” complexity in time and space for system components, slightly reduced by averaging effect of acting moments I(t). It indicates that component trajectory is more constrained as it is constantly depending on neighbouring ones, and a possible way is to represent this collective action as an average potential through a distribution function equation of which eqns(2,3) are the characteristics. Apart few existence theorems[32], study of eqns(2,3) is at very preliminary stage today. Like for eqn(1), their numerical study can be undertaken by fixing design and control parameters in F and G such that some trajectory properties are met. Here also a useful companion approach is in application of embedding theorems by fixed point methods, especially when F and G are satisfying generalized Lipschitz conditions, which often provides an embedding property in Sobolev spaces by use of substitution theorems[33]. More precise statement can be obtained when there exist upper bounding estimates in norm of the RHS of eqns(2,3). For example, if the right hand sides of eqns(2,3) satisfy inequalities of generalized Lipschitz type[34]

F ,Gat≤+ ()∑ [ btgZtcdtkZdtii () ( ())+ iii ( ()) ( ( i ()))] (4) in=1,

where Z = ||col{X,I}||, di(t) = InfX{t − hi(t,X(.))}, and the various functions gi(.), ki(.), ai(.) and bi(.) are positive in their definition intervals, there exists the following bound on the solution

⎪⎧ t ⎪⎫ Z()t≤+ΦHH−1 sgn( H ) (s )ds (5) ⎨ 0 ∫ ⎬ ⎩⎭⎪ t0 ⎪

v with H(Z) = || col{1,gi(Z),ki(Z)}|| , Φ(t) = || col{a’(t),bi(t),ci(t)di’(t)}|| , ‘= d/dt and H(v) = ∫ dv/H(v), H−1(.) its inverse function. The solution X(t) is bounded with Z(t) defined over the complete time + + −1 interval t ∈ {t0,+∞}⊂ R1 and for all Z0 = Z(t0) ∈ Rn when H (.) is bounded which generally occurs when the function is sub-linear, whereas there is a conditional bound inside the domain limited by H0 T + sgn(H)∫ Φ(s)ds = 0 in {H0,T}−space, ie in {X0,T}−space, when the function is super-linear. When p p there exists monomial bounds gi(x) = λi x , ki(x) = μi x and a(t) = 0, one obtains the explicit bound

1 dt() 1− p ⎡ ⎪⎪⎧⎫t i ⎤ Zt()≤−−⎢ Zt ( )1− p ( p 1)λμ b ( sds ) + M c() sds ⎥ (6) 0 ∑ ⎨⎬ii∫∫ ii i ⎢ in=1, ⎥ ⎣ ⎩⎭⎪⎪td00i ()t⎦

with Mi = Max{1/di’}, showing a limit in {t,Z(t0)}−plane for p>1 when the term between bracket is 0. Bounds in eqns(5,6) exhibit double advantage of being analytically meaningful and tractable, and to naturally create equivalence classes within which system dynamics are globally the same. This is exactly the expression of system natural robustness resulting from the passage of some components to complex state which can be here easily exploited. An interesting observation from such an approach is again the emphasis on importance of the manifold on which system trajectories are staying, much more than one specific trajectory which has been seen above not to contain enough information for generating an outer action to control it. Another point is the way stability is now perceived in present setting: rather than researching usual specific decay by classical Lyapounov method (for asymptotic stability), it is more generally the belonging of solution manifold to a given function space determined by its properties, see Fig.7. Obviously the bound in eqn(6) exists over the entire time interval only if the term between bracket on the right hand side of eqn(6) is > 0, which implies a relation between bounds parameters λi, μi and initial values Z(t0). So for a given initial ball || Z(t0) ||, there exists an uncertainty ball || [Δλi, Δμi] || on system parameters within which the solution belongs to a manifold M defined by the RHS of eqn(6), the embedding of which in required function space S can be studied in a second step. This opens the way to get at the same time asymptotic stability and robustness, both useful properties for control of complex systems. The approach can be extended to the opposite and more difficult situation where system uncertainty ball is larger than robustness ball[35].

5. Applications to man made systems

Evidently previous properties are of up-most importance when applied to man made systems now appearing in industry. Under economic competition, more advanced systems are conceived and worked out which include an always increasing number of heterogeneous components to be operated all together for production of higher value objects. This would imply to keep complete system mastery by efficient control, which becomes the more illusory as system dynamics cover a larger number of elements escaping from one single centralized control structure, especially by becoming complex due to higher value of coupling parameters. In parallel, to reduce system behaviour fragility, it is also interesting to reduce the number of input control parameters by transforming the system into a (partially) complex one by clustering some components into bigger parts. A trivial illustration of this approach is given by the way shepherd dogs are acting on a herd of cattle in the meadows. If there are n animals wandering around, they represent a system with 2n degrees of freedom (3n when counting their orientation). Clearly the dog understands that it is hopeless with his only 2 degrees of freedom to control all the animals, so his first action is to gather all of them in a restricted space so that by being close enough they have strong enough interactions transforming their initial complicated system into a complex one with dramatically less degrees of freedom, in fact only two like himself. Then he can control perfectly well the herd as easily observed. The astonishing fact is that dogs are knowing what to do (and they even refuse to do anything with animals unable to go into this complex stage, ie to develop gregarious potential) whereas today engineers are not yet able to proceed in similar way with their own constructions and to get corresponding benefits in terms of global mastery. This is an immense challenge industrial civilisation is facing today justifying if any the needs to study and to create these complex systems. On the representation of system in Fig.3, this would mean to vary adequate parameters to move the representative points along complex axis in order to decide exactly new system status. In any case, internal non controlled dynamics are taken care of by system self organization resulting from passing to complex state, implying that precise trajectory control is now delegated to system. The challenging difficulty is that comply with new structure, some “intelligence” has also to be delegated to the system, leading for the operator to a more supervisory position[36]. In present case, this is contribution to trajectory management by shifting usual (imposed) trajectory control to more elaborated task control [37], a way followed by all living creatures in their daily life to guarantee strong robustness while still keeping accuracy and preciseness. This illustrates if any the limited possibility of expressions from laws of Physics because they are tightly linking information flux related to the described action to power flux implied in it. Thanks to the discovery of “memory” DNA molecules, Nature has been very early able to “escape” from the constraint of solely following Physics laws strict causality, by inventing sufficiently isolated systems manipulating information flux as well (ie in storing and releasing it according to a timing fixed by survival goal). In fact resulting entropy decrease from created order cannot be maintained for ever due to their unavoidable transport and energy loss because they cannot exist in complete isolation, so Nature has been circumventing resulting isolated system finite life by inventing reproduction permitting species survival instead. In some sense, Human kind today is faced to a similar problem in the research of higher efficiency for industrial systems. After first development of tools, then to machines, and later to efficient control structure, leading to Mechatronics, a new step is now under way to give man made systems more efficiency and autonomy by delegating more “intelligence” to them, see Fig.8. This implies to search an adequate merging of information flux mastery from recent Information Technology development with power flux mastery resulting from classical long term mechanical development[38]. An elementary illustration is provided by the evolution of Robotics to comply with always more severe industrial constraints. It can be shown[14] that robotic systems are completely fixed by the 2 three parameters (Jr )i , εi and (ωdτR)i for each link i, with J the inertia moment of the link (including actuator), r the gearing ratio, ε the compliance factor and (ωdτR) the ‘’rigidity’’ factor with ωd the characteristic deformation frequency and τR the characteristic mechanical frequency. First N-link 2 robots were build up so that parameters Jr , ε and (ωdτR) <<1 for each link i. The inequalities can be shown in full generality to decouple link interactions and to keep the system rigid, so the system was in simple state. However slowness (speed is multiplied by r) could not be compensated by higher actuator rotation speed, and passing to complicated state, use of decoupling method has been allowing direct drive actuation r = 1 but with preciseness problems from inadequate control. This has been improved by developing more robust controllers in broader Mechatronics, especially to improve mastery of compliance and vibrations. Recent orientation with higher nominal parameters is toward more autonomous and more intelligent systems with larger decision delegation shifting new robots toward complex state. This new step implies the introduction of “intention” into the system and not to stay as before at simple action level of following prescribed fixed trajectory dictated by classical control. In such scheme intention is represented by the “task” communicated to the system which now determines its own trajectory. Such evolution is following the heavy trend started from beginning of human activity, as described in Fig.8.

6. Conclusion

Thorough examination of natural systems in last fifty years has confirmed the existence of systems exhibiting behaviours which do not fit with main stream scientific laws established from patient observations of Nature over past centuries. These laws are based on mechanistic representation of the Universe born from celestial body motion, ie from identifiable simple objects with well defined trajectories. With the ascent of modern technology, new natural systems have been studied and artificial ones have been constructed, both with very intricate structure implying a large number of heterogeneous components in strong interaction. Application of usual laws is often unable to describe their dynamics, because they stay outside the domain of complicated multi component systems only covered by use of reductionism method. The main reason is in the overtaking of component interaction strength which dominates enough over other effects to force the system to close on itself and to manifest an internal self-organization responsible of its new behaviour. Differently said in elementary terms, the new paradigm is that “increasing interactions between components lead to their isolation” as easily verifiable. Such systems are termed as “complex” from etymology, and their main feature is that components in complex state are internally ruled through this self-organization so that at the end they are less depending on environment action. As a consequence, natural complex systems are structurally more robust than complicated ones as evidenced by observation of living organisms, the most complex existing systems. Similarly, artificial complicated systems can be made more robust by transforming them into complex ones when linking adequately and strongly enough some of their components. As always since human origin, the price to pay for this insensitivity is the needs to replace shielded outer action by larger delegation of “intelligence” to the system for its correct internal driving. Analysis of complex systems dynamics shows, for high enough ability of system components (now called “agents”), the possibility of “emergence” of a new behaviour which is not included into the set of initial components behaviour. Though it could be the way of approaching the new global and larger “2-dimensional” picture of the Universe proposed here, to avoid trivial rephrasing of well known classical results, emergence in this context appears to be more justified when restricting its domain to action of agents teaming up for collective goal they cannot individually reach. The problem of representing complex system structure build up has been discussed, and three typical structures have been singled out. A− The broad class of reducible structure to which natural dissipative systems are belonging, usually formally amenable to be analytically projected back into complicated structure thanks to very large time and space scale difference between components. Application of asymptotic expansion methods allows to directly get important system quantities such as transport coefficients. B− The ordinary differential structure, thoroughly analyzed theoretically and by computer simulation, which shows when increasing component coupling interaction the transformation of system trajectories into finally fully chaotic and indistinguishable ones filling complete domains of state space, manifesting a layered organization of motion into equivalence classes which are the only reachable system parameters from environment. C−The most difficult delayed structure mainly due to long range component interactions, exhibiting a double complexity in space and time together, and grasping the fundamental aspect of component self determination by properly accounting for neighbouring ones within the range of interaction potential. Origin and main properties of class A have been cleared up by Physics study, but transport problem is not completely solved for continuous systems such as Fluids and Plasmas. Class B has been carefully analyzed from Mathematics, especially in the difficult case of resonances for Hamiltonian systems, but still resists to a full picture from analytical reconstitution of trajectory manifold. Class C remains mainly open as its study did not really begin yet. Functional methods approach in adapted function spaces has been shown to be successful for reducible systems in class A, in that both robustness and asymptotic stability are obtained together. Extension to classes B and C will be discussed elsewhere. As it is based on elementary source−sink description from Physics of interactions between elements of a system (the agents), and between the system and its environment, complexity paradigm discussed here can be tested through a falsifiable protocol. This bottom-up analysis is directly privileging a middle (control engineering type) approach along possible action on the system from environment in between purely (speculative) mathematical and purely (contemplative) physical classical approaches, because it fits immediately with global concepts of auto-organization and adaptive robustness resulting from transformation to complex, and allows direct application of powerful and particularly adapted modern functional analysis methods. Application to artificial man-made systems is particularly appropriate for highly performing ones, as entering complexity is opening the possibility for the “function” to feedback onto the “structure”, due to emergence produced by self-organization, a very specific property of the only living systems to date guaranteeing their remarkable survival ability. Finally, possible extension of this approach to “soft” sciences could open on orientation of economic and social sciences research towards more rigorous and potentially predictive models of human conduct.

7. References

[1] R. Descartes : Le Discours de la Méthode, Œuvres, V. Cousin, ed., Paris, 1824 [2] H.H. Pattee : Hierarchy Theory, the Challenge of Complex Systems, Braziller, New−York, 1973; J. Galbraith : Designing Complex Organisations, Addison−Wesley, Reading, MA, 1973; A. Cottrell : Emergent Properties of Complex Systems, in Encyclopedia of Ignorance, R. Duncan, M. Weston−Smith (eds.), Pergamon Press, Oxford, pp.129-135, 1977; H.A. Simon : How Complex are Complex Systems?, Philosophy of Science Assoc., Vol.2, pp.507-522, 1976; F.A. Hayek : The Theory of Complex Phenomena, in The Critical Approach to Science and Philosophy, M. Bunge (ed.), Collier McMillan, London, pp.332-349, 1964; C.C. Walker : Behaviour of a Class of Complex Systems: the Effect of Systems Size on Properties of Terminal Cycles, J. of Cybernetics, Vol.1, pp.55-67, 1971; A.E. Ferdinand : Complex Systems are Those that Work Even with Maximum Entropy, Intern. J. of General Systems, Vol.1, pp.19-33, 1974; R.J. Nelson : Structure of Complex Systems, Philosophy of Science Association, Vol.2, pp.523-542, 1976; J.L. Casti : Complexity, Connectivity and Resilience in Complex Ecosystems, IFAC Symp. on Bio and Ecosystems, Liepzig, Germany, 1977; Trappler (ed.) : Power, Autonomy, Utopia: New Approaches towards Complex Systems, Plenum Press, New York, 1986; J. Johnson : A Theory of Stars in Complex Systems, in Complexity Language and Life: Mathematical Approaches, J.L. Casti, A. Karlquist (eds.), Springer–Verlag, Berlin, pp.21-61, 1986; H. Haken : Information and Self−Organisation: a Macroscopic Approach to Complex Systems, Springer-Verlag, Berlin, 1988; H. Soodak, A.S. Iberall : Thermodynamics and Complex Systems, in Self-Organising Systems: the Emergence of Order, F.E. Yates (ed.), Plenum Press, New− York, pp.459-469, 1987; C. Dyke : The Evolutionary Dynamics of Complex Systems, OUP, New−York, 1988; S. Funtowicz, J. Ravetz : Emergent Complex Systems, Futures, Vol.26, pp.568-582, 1994; S. Wolfram : Complex Systems Theory, in Emerging Syntheses in Science, D. Pines (ed.), Addison−Wesley, Redwood City, CA, pp.183-190, 1987; H. Haken : Synergetics: An Introduction, 2nd ed., 1983, Springer Series in Synergetics, Vol.1, Springer−Verlag, Berlin, Heidelberg, New−York, 1983 and Information and Self–Organization: A Macroscopic Approach to Complex Systems, Springer Series in Synergetics, Vol.40, Springer-Verlag, Berlin, Heidelberg, New York, 1988; [3] S. Weinberg : Dreams of a Final Theory, Pantheon Books, 1993 [4] R. Laughlin : A different Universe : Reinventing Physics from the Bottom Down, Basic Books, 2005 [5] P.S. Laplace : Œuvres Complètes, Part 6, Gauthiers−Villard, Paris, 1878-1912 ; J.L. Lagrange : Mécanique Analytique, Vol.I-II, Ed. Albert Blanchard, Paris, 1788 [6] Aritotle : Physics, Part I, Chap.9 ; P.W. Anderson : More is Different, Science, Vol.177, pp.393-396, 1972 [7] M. Gell-Mann : What is Complexity?, Complexity, Vol.1, pp.16-19, 1995; G. Nicolis, C. Nicolis : Foundations of Complex Systems, World Scientific Pub., Singapore, 2007; W. Weaver : Science and Complexity, American Scientist, Vol.36, pp.536-544, 1948; G. Nicolis, I. Prigogine : Exploring Complexity : an Introduction, W.H. Freeman and Co, NY, 1989; M. Bunge : The Myth of Simplicity, Prentice−Hall, Englewood Cliffs, 1963; W.R. Ashby : Introduction to Cybernetics, Methuen, New York, 1964; M. Baldwin : Portraits of Complexity, Battelle Memorial Institute, Ohio, 1975; J.F. Traub, G.W. Wasilkowski, H. Wozniakowski : Information, Uncertainty, Complexity, Addison−Wesley, Reading, MA, 1983; J.L. Casti : Connectivity, Complexity and Catastrophe, J. Wiley & Sons, New−York, 1979; J.V. Cornacchio : System Complexity - A Bibliography, Intern. J. of General Systems, Vol.3, pp.267-271, 1977; J.N. Warfield : Planning, Policy, and Complexity, J. Wiley & Sons, New York, 1976; I. Prigogine, I. Stengers : Order out of Chaos - Man's New Dialogue with Nature, Bantam Books, Toronto, 1984; E. Grunberg : "Complexity" and "Open Systems" in Economic Discourse, J. of Economic Issues, Vol.12, pp.541-560, 1978; G.D. Brewer : Analysis of Complex Systems: An Experiment and its Implications for Policy Making, Rand Corp. Report P-4951, Santa Monica, CA, 1973; M.H. Van Emden : An Analysis of Complexity, Mathematical Centre Tracts, Mathematische Centrum Amsterdam, Amsterdam, 1971; W.R. Ashby : Some Peculiarities of Complex Systems, Cybernetic Medicine, Vol.9, pp.1-7, 1973; J.G. Kemeny : Two Measures of Complexity, the Journal of Philosophy, Vol.52, pp.722- 733, 1955; S. Winograd : Redundancy and Complexity of Logical Elements, Information and Control, Vol.5, pp.177-194, 1963; Bar−Hillel, Y. Kasher, E. Shamir : Measures of Syntactic Complexity, in Machine Translation, A.D. Booth (ed.), North−Holland, Amsterdam, pp.30-50, 1967; A. Moshowitz : Entropy and the Complexity of Graphs, Bull. Mathematical Biophysics, Vol.30, pp.175-204,225-40, 1968; R.A. Rock : Integrative Complexity and Non-Specific Set as Determinants of Performance in a Complex Solving Task, J. Behavioural Science, Vol.1, pp.17-25, 1969; A.K. Zvonkin, L.A. Levin : The Complexity of Finite Objects and the Development of the Concepts of Information and Randomness by Means of the Theory of Algorithms, Russian Math. Surveys, Vol.256, pp.83-124, 1970; S. Beer : Managing Modern Complexity, Futures, Vol.2, pp.245-257, 1970; G. Ausiello : Abstract Computational Complexity and Cycling Computations, J. Computer and System Sciences, Vol.5, pp.118-128, 1971; E. Sober : Simplicity, Clarendon Press, Oxford, 1975; W. Wimsatt : Complexity and Organisation, in Studies in the Philosophy of Sciences, K. Schaffner, R. Cohen, (eds.), Reidel, Dordretch, the Netherlands, pp.67-86, 1972; P. Cilliers : Complexity and Postmodernism, Routledge, London, 1998; J.A. Feldman : Some Decidability Results on Grammatical Inference and Complexity, Information and Control, Vol.20, pp.244-262, 1972; J.M. Seaman, F. Koenig : A Comparison Measures of Cognitive Complexity, Sociometry, Vol.37, pp.375-390, 1974; B.K. Rosen : Syntactic Complexity, Information and Control, Vol.24, pp.305-335, 1974; E. Morin : Complexity, Intern. J. of Social Science, Vol.26, pp.583-597, 1974; L. Lofgren : Complexity of Descriptions of Systems: A foundational Study, Intern. J. of General Systems, Vol.3, pp.197-214, 1974; M.A. Arbib, E.G. Manes : Foundations of System Theory: Decomposable Systems, Automatica, Vol.10, pp.285-302, 1974; H. Bremermann : Complexity of Automata, Brains and Behaviour, Lecture Notes in Bio−Mathematics, Vol.4, pp.304-331, 1974; H.W. Gottinger : Complexity and Information Technology in Dynamic Systems, Kybernetes, Vol.4, pp.129-141, 1975; N. Pippenger : Complexity Theory, Scientific American, Vol.239, pp.90-100, 1978; C. Futia : The Complexity of Economic Decision Rules, J. Mathematical Economics, Vol.4, pp.289-299, 1975; A. De Luca : Complexity and Information Theory, in Coding and Complexity, G. Longo (ed.), Springer-Verlag, Berlin, pp.207-270, 1975; T. Yokoyama : Complexity and Hierarchy Theory, Osaka Economic Papers, Vol.26, pp.26-34, 1976; H.A. Simon : The Architecture of Complexity, in the Sciences of the Artificial, MIT Press, Cambridge, Massachusetts, pp.192-229, 1981; D. Sahal : Elements of an Emerging Theory of Complexity Per Se, Cybernetica, Vol.19, pp.5-38, 1976, and : System Complexity: Its Conceptions and Measurement in the Design of Engineering Systems, IEEE Trans. on Systems, Man and Cybernetics, Vol.6, pp.440-445, 1976; S.Y. Auyang, Foundations of Complex-system theories, Cambridge University Press, 1998; J. Holland, Hidden Order : How Adaptation Builds Complexity, Helix Books, Addison Wesley, New-York, 1995; M. Waldrop, Complexity : the Emerging Science at the Edge of Order and Chaos, Touchstone, NY, 1993; J. McCabe : A Complexity Measure, IEEE Trans. on Software Engineering, Vol.2, pp. 308-320, 1976; H.W. Gottinger : Complexity and Dynamics: Applications of Dynamic Systems Theory, IEEE Trans. on Systems, Man and Cybernetics, Vol.6, pp.867-873, 1976; B.R. Gaines : System Identification, Approximation and Complexity, Intern. J. of General Systems, 3, 145-174, 1977; R. Rosen : Complexity and System Descriptions, in Systems: Approaches, Theories, Applications, W.E. Harnett ed., Reidel, Dordrecht, Holland, pp.169-175, 1977; M.J. Marcus : The Theory of Connecting Networks and their Complexity: A Review, Proc. IEEE, Vol.65, pp.1263-1271, 1977; J.A. Goguen : Complexity of Hierarchically Organised Systems, Intern. J. of General Systems, Vol.3(4), pp.233-251, 1977; B. George : Tests for System Complexity, Intern. J. of General Systems, Vol.3, pp.253-257, 1977; F.E. Yates : Complexity and the Limits to Knowledge, Amer. J. of Physiology, Vol.235, pp.R201-R204, 1978; R.J. Solomonoff : Complexity-Based Induction Systems: Comparisons and Convergence Theorems, IEEE Trans. on Information Theory, Vol.24, pp.422-432, 1978; G.J. Klir : Complexity: Some General Observations, Systems Research, Vol.2, pp.131-140, 1985; J. Pearl : On the Connection Between the Complexity and Credibility of Inferred Models, Intern. J. of General Systems, Vol.4, pp.255-264, 1978; J.E. Hopcroft, J.D. Ullman : Introduction to Automata Theory, Languages, and Computation, Addison−Wesley, Reading, Massachusetts, 1979; R.D. Levine, M. Tribus (eds.): The Maximum Entropy Principle, MIT Press, Cambridge, Massachusetts, 1979; W.B. Rouse, S.H. Rouse : Measures of Complexity of Fault Diagnosis Tasks, IEEE Trans. on Systems, Man and Cybernetics, Vol.9, pp.720-727, 1979; C. Gershenson , F. Heylighen : How Can we Think the Complex? in: Richardson, Kurt (eds.) Managing the Complex Vol. 1: Philosophy, Theory and Application (Institute for the Study of Coherence and Emergence/Information Age Publishing); F. Papentin : On Order and Complexity, J. Theoretical Biology, Vol.87, pp.1-456, 1980; P.S. Albin : The Complexity of Social Groups and Social Systems by Graph Structures, Mathematical Social Sciences, Vol.1, pp.101-129, 1980; R.L. Dussauchoy : Generalized Information Theory and the Decomposability of Systems, Intern. J. of General Systems, Vol.9, pp.13-36, 1982; P.S. Albin, H.W. Gottinger : Structure and Complexity in Economic and Social Systems, Math. Social Sciences, Vol.5, pp.253- 268, 1983; R. Hinegardner, J. Engelberg : Biological Complexity, J. Theoretical Biology, Vol.104, pp.7-20, 1983; J. Ziv : Complexity and Coherence of Sequences, in The Impact of Data Processing Techniques on Communications, H. Durand, M. di Lullo, C. Sinclair (eds.), Nijhoff, Dordrecht, pp.23-33, 1983; S. Wolfram : Universality and Complexity in Cellular Automata, Physica D, Vol.10, pp.1, 1984 and : Computational Theory of Cellular Automata, Comm. in Mathematical Physics, Vol.96, pp.15-57, 1984 and : Cellural Automaton as Models of Complexity, Nature, Vol.311, pp.419-424, 1984; Aida & al. (eds.) : The Science and Praxis of Complexity, United Nations University, Tokyo, 1984; R.E. Prather : An Axiomatic Theory of Software Complexity Research, The Computer J., Vol.27, pp.340-346, 1984; A. Nerisian : On a Definition of the Description Complexity of Finite Systems, Intern. Inst. for Applied Systems Analysis, Rept.WP-84-004, 1984; R. Rosen : Anticipatory Systems, Pergamon Press, New−York, 1985; L. Lovsz : The Mathematical Notion of Complexity, Inst. fur Okonometrie und Operations Research Rept, University of Bonn, 1985; J.K.K. Ho, D. Sculli : System Complexity and the Design of Decision Support Systems, Systems Practice, Vol.8, pp.505-516, 1995; R. Rosen : Organisms as Causal Systems which are not Mechanisms: An Essay into the Nature of Complexity, in Theoretical Biology and Complexity: Three Essays on the Natural Philosophy of Complex Systems, R. Rosen (ed.), Acad. Press, London, pp.165-203, 1985; J.L. Casti, A. Karquist (eds.) : Complexity, Language and Life: Mathematical Models, Springer-Verlag, Berlin, 1985; C. Celluci : Proof Theory and Complexity, Synthese, Vol.62, pp.173-189, 1985; J.T. Bonner : The Evolution of Complexity, Princeton Univ. Press, Princeton, NJ, 1988; C.H. Bennett : Dissipation, Information, Computational Complexity and the Definition of Organization, in Emerging Syntheses in Science, D. Pines (ed.), Addison-Wesley, Redwood City, CA, pp.215-233, 1985; R.E. Wood : Task Complexity: Definition of the Construct, Organisational Behaviour and Human Decision Processes, Vol.36, pp.60-82, 1986; B.A. Huberman, T. Hogg : Complexity and Adaptation, Physica D, Vol.22, pp.376-384, 1986; P. Grassberger : Towards a Quantitative Theory of Self- Generated Complexity, Intern. J. of Theoretical Physics, Vol.25(9), pp.907-938, 1986; C.H. Bennett : On the Nature and Origin of Complexity in Discrete, Homogenous, Locally-Interacting Systems, Foundations of Physics, Vol.16, pp.585-592, 1986; G. Kampis, V. Csanyi : Notes on Order and Complexity, J. of Theoretical Biology, Vol.124, pp.111-121, 1987; M. Koppel : Complexity, Depth and Sophistication, Complex Systems, Vol.1, pp.1087-1091, 1987; H.W. Gottinger : Choice and Complexity, Mathematical Social Sciences, Vol.14, pp.1-17, 1987; R.L. Flood, E.R. Carson : Dealing with Complexity, Plenum Press, New York, 1988; K. Mainzer : Complex Systems in Nature and Society, Physik in Unserer Zeit, Vol.28, pp.74-81, 1997 and Thinking in Complexity: The Complex Dynamics of Matter, Mind, and Mankind, 3rd. ed., Springer–Verlag, Berlin, Heidelberg, New York, 1997; H.J. Levesque : Logic and the Complexity of Reasoning, J. of Philosophical Logic, Vol.17, pp.355-389, 1988; D.J. Campbell : Task Complexity: A Review and Analysis, Academy of Management Review, Vol.13, pp.40-52, 1988; P. Turney : The Architecture of Complexity: A New Blueprint, Synthesis, Vol.79, pp.515-542, 1989; D. Stein : Lectures in the Sciences of Complexity, Addison−Wesley, Redwood City, CA, 1989; E. Berger : Computability, Complexity, Logic, North Holland, Amsterdam, 1989; R. Badii : Unfolding Complexity in Nonlinear Dynamical Systems, in Measures of Complexity and Chaos, N.B. Abraham (ed.), Plenum Press, New York, pp.313-325, 1989; P. Turney : Problems with Complexity in Gold's Paradigm of Induction-- Part I: Dynamic Complexity, Intern. J. of General Systems, Vol.17, pp.329-342, 1990; Part II: Static Complexity, Intern. J. of General Systems, Vol.17, pp.343-358, 1990; B. Rustem, K. Velupillai : Rationality, Computability and Complexity, J. of Economic Dynamics and Control, Vol.14, pp.419-432, 1990; P. Kindlmann : Stability vs. Complexity in Model Computational Communities, Lecture Notes in Bio– Mathematics, Vol.54, pp.191-207, 1984; G. d'Alessandro, A. Politi : Hierarchical Approach to Complexity with Applications to Dynamic Systems, Phys. Rev. Letters, Vol.64, pp.1609-1612, 1990; J.F. Banks, R.K. Sundaram : Repeated Games, Finite Automata and Complexity, Games and Economic Behaviour, Vol.2, pp.97-117, 1990; R. Machlin : The Complex Behaviour of Simple Machines, Physica D, Vol.42, pp.85-98, 1990; J. Gorodkin, A. Sorensen, O. Winther : Neural Networks and Cellular Automata Complexity, Complex Systems, Vol.7, pp.1-23, 1993; S. Lloyd : Physical Measures of Complexity, in 1989 Lectures in Complex Systems, E. Jen (ed.), Addison− Wesley, Reading, Mass., pp.67-73, 1990; D.A.R. George : Chaos and Complexity in Economics: Review Article, J. of Economic Surveys, Vol.4, pp.397-404, 1990; C.H. Bennett : How to Define Complexity in Physics, and Why, in Complexity, Entropy and the Physics of Information, W.H. Zurek (ed.), Addison−Wesley, Redwood City, CA, pp.137-148, 1990; J.L. Balcazar : Structural Complexity, Springer−Verlag, Berlin, 1990; G.L. Baker : Chaotic Dynamics, Cambridge Univ. Press, Cambridge, Mass., 1990; J.F. Traub, H. Wozniakowski : Information-Based Complexity: New Questions for Mathematicians, Mathematical Intelligencer, Vol.13, pp.34- 43, 1991; H. Zuse : Software Complexity, Walter de Gruyter, Berlin, 1991; K.B. Lakshmanan, S. Jayaprakash, P.K. Sinha : Properties of Control-Flow Complexity Measures, IEEE Trans. on Software Engineering, Vol.17, pp.1289-1295, 1991; D. McShea : Complexity and Evolution: What Everybody Knows, Biology and Philosophy, Vol.6(3), pp.3003-3024, 1991; G. Kampis : Self–Modifying Systems in Biology and Cognitive Science: A New Framework for Dynamics, Information and Complexity, Pergamon Press, Oxford, 1991; P. Grassberger : Information and Complexity Measures in Dynamical Systems, in Information Dynamics, H. Atmanspacher, H. Scheingraber, (eds.), Plenum Press, New−York, pp.15-33, 1991; D. Woodward : Understanding Complexity: A Critique and Synthesis, Henley Management College, Henley−on−Thames, England, 1991; H. Atmanspacher, J. Kurths, H. Scheingraber, R. Wackerbauer, A. Witt : Complexity and Meaning in Nonlinear Dynamical Systems, Open systems and Information Dynamics, Vol.1, pp.269-289, 1991; F. Heylighen : Self–Organization, Emergence and the Architecture of Complexity, in Proc. European Congress on System Science, AFCET (eds.), Paris, pp.23-32, 1992; J.P. Crutchfield : Knowledge and Meaning ... Chaos and Complexity, in Modeling Complex Phenomena, L. Lam, V. Naroditsky (eds.), Springer, Berlin, pp.66-101, 1992; J.L. Casti : Reality Rules: Part I. Picturing the World in Mathematics: The Fundamentals, Part II. Picturing the World in Mathematics: The Frontier, J. Wiley & Sons, New−York, 1992, and : The Simply Complex: Trendy Buzzword or Emerging New Science, Bull. of the Santa Fe Institute, Vol.7, pp.10-13, 1992; R. Badii : Complexity and Unpredictible Scaling of Hierarchical Structures, in Chaotic Dynamics, Theory and Practice, T. Bountis (ed.), Plenum Press, New−York and London, pp.1-19, 1992; S. Kauffman : At Home in the Universe: The Search for Laws of Self–Organization and Complexity, Oxford Univ. Press, New−York, 1995; P.M. Auger : Hierarchically Organized Systems − The Response to Complexity, J. of Scientific & Industrial Research, Vol.51, pp.725-735, 1992; R. Ruthen : Adapting to Complexity, Scientific American, Vol.268, pp.130-135, 1993; S.A. Kauffman : The Origins of Order, Oxford Univ. Press, New York, 1993; J.E. Bates, H.K. Shepard : Measuring Complexity Using Information Fluctuation, Phys. Letters A, Vol.172, pp.416-425, 1993; H.J. Holm : Complexity in Economic Theory, University of Lund: Lund Economic Studies, Lund, Sweden, 1993; W.B. Arthur : Why Do Things Become More Complex?, Scientific American, Vol.268, pp.92, 1993; F.E. Yates : Order and Complexity in Dynamical Systems − Homeodynamics as a Generalised Mechanics for Biology, Mathematical and Computer Modelling, Vol.19, pp.49-74, 1994; T.S. Ray : Evolution, Complexity, Entropy and Artificial Reality, Physica D, Vol.75, pp.239-263, 1994 and : Evolution and Complexity, in Complexity: Metaphors, Models and Reality, G. Cowan, D. Pines, D. Meltzer (eds.), Addison−Wesley, Reading, Mass., pp.161-173, 1994; A. Reggiani, P. Nijkamp : Competition and Complexity in Spatially Connected Networks, System Dynamics Review, Vol.11, pp.51-66, 1995; L. Poon, C. Grebogi : Controlling Complexity, Phys. Rev. Letters, Vol.75, pp.4023-4026, 1995; J. Horgan : From Complexity to Perplexity, Scientific American, Vol.272, pp.74-79, 1995; R. Badii, A. Politi : Complexity: Hierarchical Structures and Scaling in Physics, Cambridge Univ. Press, Cambridge, 1997; E.L. Khalil : Organizations, Naturalism and Complexity, Review of Social Economy, Vol.53, pp.393-419, 1995; A.V. Zeigarnik, O.N. Temkin : A Graph−Theoretic Model of Complex Reaction Mechanisms: a New Complexity Index for Reaction Mechanisms, Kinetics and Catalysis, Vol.37, pp.372-385, 1996; D. Helbing, F. Schweitzer (Guest eds.) : Complex Dynamics in Economics, Advances in Complex Systems, Vol.4(1), pp.1-176, 2001; J.L. Casti : Connectivity, Complexity and Catastrophe in Large-Scale Systems, Intern. Ser. on Applied Systems Analysis, Vol.7, J. Wiley & Sons, Chichester, New York, 1979; D.P. Feldman, J.P. Crutchfield : Measures of Statistical Complexity: Why? www.santafe.edu/~cmg/papers/mscw.pdf , Nov. 11, 1997; C.H. Bennett : How to Define Complexity in Physics and Why, in From Complexity To Life: On the Emergence of Life and Meaning, N.H. Gregersen (ed.), Oxford Univ. Press, New York, 2003; R. Lòpez-Ruiz, H. L. Mancini, X. Calbet : A Statistical Measure of Complexity, Phys. Lett A, Vol.209, pp.321-326, 1995; R. Feistel, W. Ebeling : Evolution of Complex Systems: Self–Organisation, Entropy and Development, Mathematics and its Applications, East European Series, Vol.30, Kluwer Acad. Publ., Dordrecht, Boston, London, 1989; K. Frenken : A Complexity Approach to Innovation Networks: The Case of the Aircraft Industry (1909-1997), Research Policy – A Journal Devoted to Research Policy, Research Management and Planning, Vol.29, pp.257-272, 2000; R. Lòpez−Ruiz : Shannon Information, LMC Complexity and Rènyi Entropies: a Straightforward Approach, http://arxiv.org/PS_cache/nlin/pdf/0312/0312056 .pdf , 2003; X. Calbet, R. Lòpez–Ruiz : Tendency towards Maximum Complexity in an Isolated Non-equilibrium System, Phys. Rev. E, Vol.60, 6-6116(9), 2001; J.S. Shiner, M. Davison, P.T. Landsberg : Simple Measure for Complexity, Phys. Rev. E, Vol.59, pp.1459-1464, 1999; J.P. Crutchfield , D.P. Feldman, C. R. Shalizi : Comment on ‘Simple Measure for Complexity, Phys. Rev. E, Vol.62, pp.2559-2564, 2000; J. Shallit, W. Elsberry : Playing Games with Probability: Dembski’s Complex Specified Information, in M. Young, T. Edis (eds), Why Intelligent Design Fails: A Scientific Critique of the New Creationism, Piscataway, N.J., 2004; G. Nicolis, I. Prigogine : Exploring Complexity : an Introduction, W.H. Freeman and Co, New−York, 1989 [8] R. Albert, A.L. Barabasi : Topology of Evolving Networks: Local Events and Universality, Phys. Rev. Letters, Vol.85, pp.5234-5237, 2000; also Statistical Mechanics of Complex Networks, Reviews of Modern Physics, Vol.74, pp.47-97, 2002; R. Albert, H. Jeong, A.L. Barabasi : Diameter of the World−Wide Web, Nature, London, Vol.401, pp.130-131, 1999; also Error and Attack Tolerance of Complex Networks, Nature, London, Vol.406, pp.378-382, and 409-542 (correction), 2000; A.L. Barabasi : Linked: The New Science of Networks, Perseus Publishing, Cambridge, MA, 2002; – with H. Jeong, Z. Néda, E. Ravasz, A. Schubert, T. Vicsek : Evolution of the Social Network of Scientific Collaborations, Physica, Vol.A-311, pp.590-614, 2002; A.F. Bartholomay : G. Bianconi, A.L. Barabasi : Competition and Multi–Scaling in Evolving Networks, Europhysics Letters, Vol.54, pp.436-442, 2001, also : Bose–Einstein Condensation in Complex Networks, Phys. Rev. Letters, Vol. 86, pp.5632-5635, 2001; S. Bornholdt, H.G. Schuster (eds.) : Handbook of Graphs and Networks : From the Genome to the Internet, Wiley−VCH Verlag, Weinheim, 2003; A. Broder, R. Kumar, F. Maghoul, P. Raghavan, S. Rajagopalan, R. Stata, A. Tomkins, J. Wiener : Graph Structure in the Web, Computer Networks – the Intern. J. of Computer and Telecom. Networking , Vol.33, pp.309-320, 2001 – See also special issue: Proc. 9th. Intern. World Wide Web Conference, the Web: the Next Generation, Amsterdam, May 15-19, 2000, Computer Networks – the Intern. J. of Computer and Telecom. Networking, Vol.33(1-6), v- viii, pp.1-843, 2000; M. Buchanan : Nexus: Small Worlds and the Groundbreaking Science of Networks, W. W. Norton & Company, New−York, London, 2002; S.N. Dorogovtsev, J.F.F. Mendes, A.N. Samukhin : Principles of Statistical Mechanics of Uncorrelated Random Networks, Nucl. Phys. B, Vol.666, pp.396-416, 2003; B. Brossel, A.J. McKane : Modelling Food Webs in Handbook of Graphs and Networks: From the Genome to the Internet, S. Bornholdt, H.G. Schuster (eds.), Wiley−VCH Verlag, Weinheim, pp. 218-247, 2003; W. Ebeling, I. Sonntag, L. Schimansky-Geier : On the Evolution of Biological Macromolecules II: Catalytic Networks, Studia Biophysica, Vol.84, pp.87-88, 1981; P. Erdös, A. Rényi : On Random Graphs I. Publicationes Mathematicae (Debrecen) Vol.6, pp.290-297, 1959 and : On the Evolution of Random Graphs, Publications of the Math. Institute of the Hungarian Academy of Sciences, Vol.5, pp.17-61, 1960; M. Faloutsos, P. Faloutsos, C. Faloutsos : On Power–Law Relationships of the Internet Topology, Computer Communications Review, Vol.29, pp.251-262, 1999; H.I. Sonntag : Analysis of the Selection Equations for a Multivariable Population Model : Deterministic and Stochastic Solutions and Discussion of the Approach for Populations of Self- Reproducing Biochemical Networks, J. Theoretical Biology, Vol.93, pp.325-361, 1981; B.A.Huberman : the Laws of the Web: Patterns in the Ecology of Information, The MIT Press, Cambridge, MA, 2001; S. Jain, S. Krishna : A Model for the Emergence of Cooperation, Interdependence and Structure in Evolving Networks, Proc. National Acad. Sciences of the U.S.A., Vol.98, pp.543-547, 2001; H. Jeong, B. Tombor, R. Albert, Z.N. Oltvai, A.L. Barabasi : The Large–Scale Organization of Metabolic Networks, Nature (London), Vol.407, pp.651-654, 2000; A. Kirman : Economic Networks, in Handbook of Graphs and Networks: From the Genome to the Internet, S. Bornholdt, H.G. Schuster (eds.), Wiley–VCH Verlag, Weinheim, pp. 273-294, 2003; Yu.L. Klimontovich : Statistical Theory of Open Systems. Part 1: A Unified Approach to Kinetic Description of Processes in Active Systems, Fundamental Theories of Physics, Vol.67, Kluwer Acad. Publishers, Dordrecht, Boston, London, 1995; L. Leydesdorff : A Sociological Theory of Communication: The Self−Organization of the Knowledge−Based Society, Universal Publishers, Parkland, FL, 2nd. ed. 2003; S.M. Maurer, B.A. Huberman : Competitive Dynamics of Web Sites, J. Economic Dynamics and Control, Vol.27, pp.2195-2206, 2000; M. Juillard, A. Marcet (guest eds.) : special issue Computing in Economics and Finance, Proc. 6th Intern. Conf. of the Society for Computational Economics, Barcelona, Spain, July 6-8, 2000, J. Economic Dynamics and Control, Vol.27(11-12), pp.1939-2265, 2000; S. Milgram : the Small World Problem, Psychology Today, Vol.1, pp.60-67, 1967; J.M. Montoya, R.V. Solé : Small World Patterns in Food Webs, J. Theoretical Biology, Vol.214, pp.405-412, 2002; C. Nachtigall : Selbstorganisation und Gewalt. Internationale Hochschulschriften, Vol.259, Waxmann Verlag, New York, München, Berlin, 1998; M.E.J. Newman : Scientific Collaboration Networks: Part I. Network Construction and Fundamental Results. Phys. Rev. E, Vol.64, 016131, 8 pp., 2001; Part II. Shortest Paths, Weighted Networks, and Centrality. Phys. Rev. E, Vol.64, 016132, 7 pp., 2001, and Models of the Small World, J. Statistical Physics, Vol.101, pp.819-841, 2000; – with D.J. Watts : Scaling and Percolation in the Small-World Network Model, Phys. Rev. E, Vol.60, pp.7332-7342, 1999; A. Pyka, G. Küppers (eds.) : Innovation Networks: Theory and Practice, New Horizons in the Economics of Innovation Series, Edward Elgar Publ., Cheltenham, Northampton, MA, 2002; I. Rechenberg : Evolutionsstrategie ’94, Werkstatt Bionik und Evolutionstechnik, Vol.1, Frommann-Holzboog-Verlag, Stuttgart, Bad Cannstatt, 1994; A. Scharnhorst : Complex Networks and the Web: Insights from Nonlinear Physics. J. Computer Mediated Communication, Vol.8(4), 2003, http://www.ascusc.org/jcmc/vol8/issue4/scharnhorst.html ; H.P. Schwefel, W. Ebeling, H. Rosé, O. Weiss : Optimization of Road Networks Using Evolutionary Strategies, Evolutionary Computation, Vol.5, pp.419-438, 1997; J. Scott : Social Network Analysis: A Handbook, 2nd ed., SAGE Publ., London, Thousands Oaks, CA, 2000; 15. R.V. Solé, J.M. Montoya : Complexity and Fragility in Ecological Networks, Proc. Royal Society London, Series B – Biological Sciences, Vol.268, pp.2039-2045, 2001; I. Sonntag : Random Networks of Catalytic Biochemical Reactions, Biometrical J., Vol.26, pp.799-807, 1984, and : Application of the Percolation Theory to Random Networks of Biochemical Reactions, Biometrical Journal, Vol.26, pp.809-813, 1984; – with R. Feistel, W. Ebeling : Random Networks of Catalytic Biochemical Reactions, Biometrical J., Vol.23, pp.501-515, 1981; S.H. Strogatz : Exploring Complex Networks, Nature (London), Vol.410, pp.268-276, 2001; O.N. Temkin, A.V. Zeigarnik, D. Bonchev : Chemical Reaction Networks: A Graph−Theoretical Approach, CRC Press, Boca Raton, FL, 1996; A. Wagner, D. Fell : The Small World Inside Large Metabolic Networks, Proc Royal Society London, Series B – Biological Sciences, Vol.268, pp.1803-1810, 2000; D.J. Watts : Small Worlds: The Dynamics of Networks between Order and Randomness, Princeton Studies in Complexity, Princeton Univ. Press, Princeton, NJ, 1999; – with S.H. Strogatz : Collective Dynamics of ‘Small World’ Networks, Nature (London), Vol.393, pp.440-442, 1998; D. Armbruster, K. Kaneko, A.S. Mikhailov : Networks of Interacting Machines, World Scientific Lecture Notes in Complex Systems, Vol.3, Singapore, 2005 [9] D. McShea : Metazoan Complexity and Evolution: is There a Trend?, Evolution, Vol.50, pp.477-492, 1996; P.T. Saunders, M.W. Ho : On the Increase of Complexity in Evolution, J. Theoretical Biology, Vol.90, pp.515- 530, 1981; B. Goodwin : How the Leopard Changed Its Spots: The Evolution of Complexity, Weidenfield & Nicholson, London, 1994 : B. Edmonds : What is Complexity? the Philosophy of Complexity per se with Application to Some Examples in Evolution, The Evolution of Complexity, F. Heylighen & Aerts (eds.), Kluwer, Dordrecht; R. Steward : Self−Organization in Evolution : a Mathematical Perspective, Phil. Trans. Royal Soc. (London), Vol.A361, pp.1101-1123, 2003; N.H. Gregersen (ed.) : From Complexity to Life: On the Emergence of Life and Meaning, Oxford Univ. Press, New−York, 2003; F. Heylighen : The Growth of Structural and Functional Complexity during Evolution, The Evolution of Complexity, Kluwer Acad. Publ., 1996; W. Ebeling, M.A. Jimenez−Montano : On Grammars, Complexity and Information Measures of Biological Macromolecules, J. Math. Bioscience, Vol.52, pp.3-71, 1980; B. Cohen, W.T. Harwood, M.I. Jackson : The Specification of Complex Systems, Addison Wesley, New−York, 1986; R.L. Flood, E.R. Carson : Dealing with Complexity, Plenum Press, New−York, 2nd ed., 1993; S. Puri, S. Datta : Nonlinearities in Complex Systems, Norosa Publ. House, Indian Institute of Advanced Studies, 1997 [10] I. Prigogine : Introduction to Thermodynamics of Irreversible Processes, American Lecture Series, Vol. 185, Ch.C. Thomas Publisher, Springfield, IL., 3rd ed., Interscience Publ., New−York, 1967: – with G. Nicolis : Self−Organization in Non−Equilibrium Systems: From Dissipative Structures to Order through Fluctuations, J. Wiley & Sons, New−York, 1977; Yu.L. Klimontovitch : Theory of Open Systems, Kluwer, Dordrecht, 1995; A. Kleidon, R.D. Lorenz (Eds.), Non-equilibrium thermodynamics and the production of entropy : life, earth, and beyond, Springer, 2005 [11] B.B Kadomtsev : Self−Organization and Transport in Tokamak Plasma, Plasma Phys. and Nucl. Fus., Vol.34(13), pp.1931-1938, 1992; P. Bautay, I.M. Janosi : Self−Organization and Anomalous Diffusion, Physica A, Vol.185(1-4), p.11-18,1992; C.K. Biebricher, G. Nicolis, P. Schuster : Self−Organization in the Physico− Chemical and Life Sciences, Vol.16546, EU Report, 1995; F. Schweitzer : Multi−Agent Approach to the Self− Organization of Networks, in Understanding and Managing Complex Agent−Based Dynamical Networks, F. Reed−Tsochas, N.F. Johnson, J. Efstathiou, eds., World Scientific, Singapore, 2007; W. Ebeling, F. Schweitzer : Self-Organization, Active Brownian Dynamics, and Biological Applications, Nova Acta Leopoldina NF Vol.88(332), pp.169-188, 2003 [12] P.I. Strand : Predictive Simulations of Transport in Tokamaks, PhD Thesis, Dept of Electromagnetics, Chalmers Univ. of Technology, 1999; F. Perkins and al. : Nondimensional Transport Scaling in the Tokamak Fusion Test Reactor : is Tokamak Transport Bohm or Gyro−Bohm?, Phys. Fluids B, Vol.5, pp.422-498, 1998; A.J. Redd, A.H. Kritz, G. Bateman, J.E. Kinsey : Sensitivity of Predictive Tokamak Plasma Transport Simulations, Phys. of Plasmas, Vol.4, pp.2207-2214, 1997 [13] V.E. Zakharov, V.S. L’vov, G. Falkovich : Kolmogorov Spectra of Turbulence, Springer−Verlag, New− York, 1992; M. Farge, K. Schneider, G. Pellegrino, A.A. Wray, R.S. Rogallo : Coherent Vortex Extraction in 3D Homogeneous Turbulence : Comparison between CVS-wavelet and POD-Fourier Decompositions, Phys. Fluids, Vol.15(10), pp.2886-2896, 2003; R. Temam : the Navier–Stokes Equations, North Holland, Amsterdam, 1978; Yu.L. Klimontovich : Turbulent Motion and Structure of Chaos, Kluwer, Dordrecht, 1991; U. Frisch : Turbulence, Cambridge Univ. Press, Mass., 1995; B. Nikolaenko, C. Foias, R. Temam, The Connection between Infinite Dimensional and Finite Dimensional Dynamical Systems, AMS, 1992 [14] M. Cotsaftis : Lectures on Advanced Dynamics, Taiwan Univ., Taipeh, R.O.C., 1993 [15] R.K. Standish : On Complexity and Emergence, Complexity International, Vol.9, pp.1-6, 2004; J. Fromm, The Emergence of Complexity, Kassel Univ. Press, 2004; F. Schweitzer : Self−Organization of Complex Structures: From Individual to Collective Dynamics, Gordon and Breach Science Publ., Amsterdam, 1997; S. Johnson : Emergence, Penguin Books, New–York, 2001; M.A. Bedau : Weak Emergence, Philosophical Perspectives: Mind, Causation, and World, J. Tomberlin (Ed.), Vol.11, Blackwell pp.375-399, 1997; S. Strogatz, Sync: The Emerging Science of Spontaneous Order, 2003, Theia; G. Di Marzo Serugendo : Engineering Emergent Behaviour: a Vision, Proc. Multi−Agent Based Simulation III, 4th Intern. Workshop, MABS 2003 Melbourne, Australia, July 2003, David Hales et al. (eds), LNAI 2927, Springer, 2003; J. Ferber, O. Gutknecht, F. Michel : From Agents To Organizations: An Organizational View of Multi−Agent Systems, Agent−Oriented Software Engineering IV, AOSE 2003, Paolo Giorgini et al. (Eds.), LNCS 2935, Springer, pp.214-230, 2004; C. Bernon, M.-P. Gleizes, S. Peyruqueou, G. Picard : ADELFE, a Methodology for Adaptive Multi-Agent Systems Engineering, 3rd Intern. Workshop on "Engineering Societies in the Agents World III" (ESAW 2002), Petta, Tolksdorf & Zambonelli (Eds.) LNCS 2577, Springer pp.156-169, 2003; J. Fromm : Types and Forms of Emergence, http://arxiv.org/abs/nlin.AO/0506028; P. Bresciani, A. Perini, P. Giorgini, F. Giunchiglia, J. Mylopoulos : A Knowledge Level Software Engineering Methodology for Agent Oriented Programming, Proc. 5th Intern. Conf. on Autonomous Agents, ACM Press pp.648–655, 2001; J. Lind : Iterative Software Engineering for Multi–Agent Systems–the MASSIVE Method, LNCS 1994, Springer, 2001; N. Jennings : On Agent Based Software Engineering, Artificial Intelligence, Vol.117(2) pp.277-296, 2000; J.O. Kephart, Research Challenges of Autonomic Computing, Proc. ICSE 05 pp.15-22, 2005;; M. Kolp, P. Giorgini, J. Mylopoulos : A Goal−Based Organizational Perspective on Multi–Agent Architectures, Intelligent Agents VIII: Agent Theories, Architectures, and Languages, LNAI 2333, Springer pp.128–140, 2002; P. Maes : Modelling Adaptive Autonomous Agents, http://citeseer.ist.psu.edu/42923.html; K.P. Sycara : Multi−Agent Systems, AI Magazine, Vol.19(2) (1998) 79- 92; R. Abbott : Emergence Explained, to be published, see also http//cs.calstatela.edu/~wiki/~images/9/90/ Emergence.Explained.pdf ; D. Yamins : Towards a Theory of "Local to Global" in Distributed Multi-Agent Systems, Proc. Intern. Joint Conf. on Autonomous Agents and Multi Agent Systems (AAMAS 2005); W. Vogels, R. van Renesse, K. Birman : The Power of Epidemics: Robust Communication for Large−Scale Distributed Systems, ACM SIGCOMM Computer Communication Review, Vol.33(1), pp.131-135, 2003; F. Zambonelli, N.R. Jennings, M. Wooldridge : Developing Multi−Agent Systems: The Gaia Methodology, ACM Trans. on Software Engineering and Methodology, Vol.12(3), pp.317-370, 2003 [16] L. Schimansky−Geier, T. Pöschel (eds.) : Stochastic Dynamics, Springer, Berlin, 1997; J. Doob : Stochastic Processes, J. Wiley & Sons, New−York, 1953; D.H.E. Gross : Micro−Canonical Thermodynamics, World Scientific, Singapore, 2001; D.N. Zubarev, G. Röpke, A.A. Morozov : Statistical Mechanics of Non−Equilibrium Processes, V.C.H. Viley, Weinheim, Berlin, Vol.I, 1996, Vol.II, 1997; R. Yulmetyev, P. Hänggi, F. Gafarov : Statistical Theory of Non−Stationary Time Correlation in Complex Systems with Discrete Time, Nonlinear Phenomena in Complex Systems, Vol.5(2), pp.129,136, 2002 [17] V.I. Arnold, A. Avez : Ergodic Problems of Classical Mechanics, Advanced Book Classics, Addison− Wesley, 1989; R. L. Jones, I. Ostrovskii, J. Rosenblatt : Square Functions in Ergodic Theory, Ergodic Theory Dynamical Systems Vol.16, pp.267-305, 1996; A.G. Kachurovskii : The Rate of Convergence in Ergodic Theorems, Russian Math. Surveys Vol.51(4), pp.653-703, 1996; Ya.G. Sinaï : Introduction to Ergodic Theory, Princeton Univ. Press, Princeton, NJ, 1976; I.P. Cornfeld, S.V. Fomin, Y.G. Sinai : Ergodic Theory, Springer- Verlag, Berlin, 1982; Karl Petersen : Ergodic Theory, Cambridge Studies in Advanced Math., Cambridge Univ. Press, Cambridge, Mass., 1983; Y. Pesin, L. Barreira : Lyapunov Exponents and Smooth Ergodic Theory, University Lecture Series n°23, American Math. Soc., Providence, RI, 2001; T. Bedford, M. Keane (eds.) : Ergodic Theory, Symbolic Dynamics and Hyperbolic Spaces, Oxford Univ. Press, Oxford, 1991; J.M. Ollagnier : Ergodic Theory and Statistical Mechanics, Lecture Notes in Math. N°1115, Springer-Verlag, Berlin, 1985; J. Lebowitz, O. Penrose : Modern Ergodic Theory, Phys. Today, Vol.26, pp.155-175, 1973; D. Ruelle : Ergodic Theory of Differentiable Dynamical Systems, Public. Mathématiques IHÉS, Vol.50, pp.27-58, 1979; C. Pugh, M. Shub : Stable Ergodicity, Bull. Amer. Math. Soc., Vol.41, pp.1-41, 2004; G.W. Mackey : Ergodic Theory and its Significance for Statistical Mechanics and Probability Theory, Advances in Mathematics, Vol.12(2), pp.178-268, 1974; G.D. Birkhoff ; Proof of the Ergodic Theorem, Proc. National Acad. Sc. USA, Vol.17 pp.656-660, 1931; M. Kac : Probability and Related Topics in Physical Science, Lectures in Applied Math. Series, Vol.1a, Amer. Math. Soc., 1957; O. Penrose : Foundations of Statistical Mechanics, Rept Progress Physics, Vol.42, pp.1937- 2006, 1979 [18] S. Camazine, J.L. Deneubourg, N.R. Franks, J. Sneyd, G. Theraulaz, E. Bonabeau, Self−Organization in Biological Systems, Princeton Univ. Press, Princeton, New-Jersey, 2002; Wei-Min Shen et al. : Hormone− Inspired Self−Organization and Distributed Control of Robotic Swarms, Autonomous Robots, Vol.17, pp.93-105, 2004; E. Bonabeau, M. Dorigo, G. Theraulaz : Swarm Intelligence : from Natural to Artificial Systems, Oxford Univ. Press, New−York, 1999 [19] M. Eigen : The Self–Organization of Matter and the Evolution of Biological Macromolecules, Die Natur– wissenschaften, Vol.58, pp.465-523, 1971; – with P. Schuster : The Hypercycle: A Principle of Natural Self– Organization. Part A: Emergence of the Hypercycle, Die Naturwissenschaften, Vol.64, pp.541-565, 1977, and Part C: The Realistic Hypercycle, Die Naturwissenschaften, Vol.65, pp.341-369, 1978; F.E. Yates (ed.) : Self− Organizing Systems: The Emergence of Order , Plenum Press, 1987; R. Lewin : Complexity−Life at the Edge of Chaos, Macmillan, 1993; A.Y. Lerner (ed.) : Principles of Self−Organization, Mir, Moscow, 1966; S. Nolfi, D. Floreano : Evolutionary Robotics : the Biology, Intelligence and Technology of Self–Organizing Machines, The MIT Press, Cambridge, Mass., 2000; R. Serra, M. Andretta, M. Compiani, G. Zanarini : Introduction to the Physics of Complex Systems (the Mesoscopic Approach to Fluctuations, Nonlinearity and Self–Organization), Pergamon Press, 1986 [20] W. Weidlich : Socio−Dynamics: A Systematic Approach to Mathematical Modelling in the Social Sciences, Harwood Academic Publishers, Amsterdam, 2000; – with G. Haag : Concepts and Models of Quantitative Sociology: The Dynamics of Interacting Populations, Springer Series in Synergetics, Vol.14. Springer-Verlag, Berlin, Heidelberg, New−York, 1983; S. Wolfram : A New Kind of Science, Wolfram Media, 2002; D. Bohm, B.J. Hiley : The Undivided Universe, Routledge, New–York, 1993; A. Bellouquid, M. Delitala : Mathematical Modelling of Complex Biological Systems – a Kinetic Theory Approach, Birkhauser, Boston, 2006 [21] S. Chandrasekhar : Stochastic Problems in Physics and Astronomy, Rev. Mod. Phys., Vol.15(1), pp.1-89, 1943; H. Risken : The Fokker-Planck Equation (2nd. ed.), Springer−Verlag, Berlin, 1996 [22] M. Cotsaftis : A Passage to Complex Systems, Proc. CoSSoM’2006, Sept. 24-27, Toulouse, France, 2006, to be published in Springer Series “Understanding Complex Systems”

[23] D. Salamon, The Kolmogorov-Arnold-Moser Theorem, Preprint, ETH-Zürich, 1986; D. Salamon, E. Zehnder : KAM Theory in Configuration Space, Comment. Math. Helv., Vol.64, pp.84–132, 1989; M. Sevryuk : KAM-Stable Hamiltonians, J. Dyn. Control Syst., Vol.1, pp. 351–366, 1995; J. Pöschel : A Lecture on the Classical KAM Theorem, Proc. Symp. Pure Math., Vol.69, pp.707-732, 2000; R.S. Hamilton : The Inverse Function Theorem of Nash Moser, Bull. A.M.S., Vol.7, pp.65-222, 1982; S. Alinhac, P. Gerard : Opérateurs Pseudo-Différentiels et Théorème de Nash-Moser, Inter-Edition et Edition du CNRS,1991; V.I. Arnold : Mathematical Methods of Classical Mechanics, 2nd Ed., Springer Verlag, 1989; also : Dynamical Systems, Encyclopaedia of Math. Sci., Vol.3, Springer Verlag, 1988; J. Moser : On Invariant Curves of Area Preserving Mappings of an Annulus, Nachr. Akad. Wiss. Gött., Math. Phys. Kl, pp.1-20, 1962; L. Markus, K. Meyer : Generic Hamiltonian Systems are neither Integrable nor Ergodic, Mem. Am. Math. Soc., Vol.144, 1974; J. Moser : Convergent Series Expansions for Quasi-periodic Motions, Math. Ann., Vol.169, pp.136-176, 1967; also : Stable and Random Motions in Dynamical Systems, Princeton Univ. Press, 1973; – with E. Zehnder : Lectures on Hamiltonian Systems, Lecture Notes, ITCP, 1991 [24] M. Tabor : Chaos and Integrability in Nonlinear Dynamics, an Introduction, Wiley and Sons, New-York, 1989; I. Stewart : Does God Play Dice? The New Mathematics of Chaos, Penguin, Harmondsworth, 1997; N.B. Abraham, A.M. Albano, N.B. Tufillaro : Complexity and Chaos, in Measures of Complexity and Chaos, N.B. Abraham (ed.), Plenum Press, New Jersey, pp.1-25, 1989; K.T. Alliwood, T.D. Sauer, J.A.Yorke : Chaos. An Introduction to Dynamical Systems, Springer, New–York, 1997; F.C. Moon : Chaotic Vibrations, Wiley- Interscience, New–York, 1987; E. Ott : Chaos in Dynamical Systems, Cambridge Univ. Press, Cambridge, Mass., 1993; R.C. Hilborn : Chaos and Nonlinear Dynamics, Oxford Univ. Press, Oxford, UK, 1994; I. Prigogine : Les Lois du Chaos, Nouvelle Bibliothèque Scientifique, Flammarion, Paris, 1993; G. Chen : Controlling Chaos and Bifurcations in Engineering Systems, CRC Press, Boca Raton, Florida, 2000; S. Wiggins : Chaotic Transport in Dynamical Systems, Springer–Verlag, New−York, 1991; S.H. Strogatz : Nonlinear Dynamics and Chaos with Applications to Physics, Biology, Chemistry, and Engineering, Addison−Wesley, Reading, Mass., 1994; also : Sync: The Emerging Science of Spontaneous Order, Theia, 2003; P. Cvitanovic : Universality in Chaos, Adam Hilger, Bristol, 1984; R.C. Hilborn : Chaos and Nonlinear Dynamics, Oxford Univ. Press, Oxford, UK, 1994; N.A. Magnitskii, S.V. Vasilevich : New Methods for Chaotic Dynamics, World Scientific Series on Nonlinear Science, Ser.A, Singapore, 2006; T. Kapitaniak : Chaos for Engineers : Theory, Applications and Control, Springer, Berlin, 1998; R.Z. Sagdeev, D.A. Usikov, G.M. Zaslavsky : Nonlinear Physics ; from Pendulum to Turbulence and Chaos, Harwood Academic, New−York, 1988 [25] H. Poincaré, Les Méthodes Nouvelles de la Mécanique Céleste, 3 Vol., Gauthier−Villars, Paris, 1892-1899; V.I. Arnold : Small Denominators and Problems of Stability of motion in Classical and Celestial Mechanics, Russian Math. Surveys, Vol.18(6), pp.85-191,1963; V.I. Arnold : Mathematical Methods of Classical Mechanics, Springer−Verlag, New−York, 1978; V.I. Arnold,V.V. Koslov, A.I. Neishtadt : Dynamic Systems III : Mathematical Aspects of Classical and Celestial Mechanics, Springer Verlag, Berlin, 1993; D. Bambusi : Nekhoroshev Theorem for Small Amplitude Solutions in Nonlinear Schrdinger Equations, Math. Z., Vol.230(2), pp.345-387, 1999; G. Benettin, G. Gallavotti : Stability of Motion near Resonances in Quasi−Integrable Hamiltonian Systems, J. of Statistical Physics, Vol.44(3-4), pp.293-338, 1986; E. Bierstone, P.D. Milman : Semi Analytic and Sub Analytic Sets, Publ. Math. IHES, Vol.67, pp.5-42, 1988; H.W. Broer, G.B. Huitema, M.B. Sevryuk : Quasi−Periodicity in Families of Dynamical Systems : Order amidst Chaos, LNM 1645, Springer− Verlag, New−York, 1996; A. Katok, B. Hasselblatt : Introduction to the Modern Theory of Dynamical Systems, Cambridge Univ. Press, Mass., 1996; N.N. Bogoliubov, Yu.A. Mitropolskii, A.M. Samoilenko : Methods of Accelerated Convergence in Nonlinear Mechanics, Springer−Verlag, Berlin, 1976 [26] M. Hirsch, C. Pugh, M. Shub : Invariant Manifolds, Lecture Notes in Math. 583, Springer−Verlag, Berlin, 1977; M. Tabor : Chaos and Integrability in Nonlinear Dynamics, an Introduction, Wiley and Sons, New−York, 1989; A. Goriely : Integrability and non Integrability of Dynamical Systems, World Scientific Publ., 2001 [27] S. Lefschetz : Stability of Nonlinear Control Systems, Academic Press, NY, 1965; M.A. Aizerman, F.R. Gantmacher : Absolute Stability of Regulator Systems, Holden−Day, San Franscisco, 1964; G.E. Dullerud, F. Paganini : A Course in Robust Control Theory : a Convex Approach, Springer−Verlag, New−York, 2000; J.C.S. Hsu, A.U. Meyer : Modern Control Principles and Applications, McGraw−Hill, New−York, 1968; S. Arimoto : Control Theory of Nonlinear Mechanical Systems : a Passivity Based and Circuit Theoretic Approach, Oxford Univ. Press, Oxford, UK, 1996; S. Russell, P. Norvig : Artificial Intelligence : a Modern Approach,

Prentice−Hall, 1995; J.H. Burl : Linear Optimal Control, H2 and H∞ Methods, Addison-Wesley Longman, Menlo Park, CA, 1999; G.A. Leonov, I.V. Ponomarenko, V.B. Smirnova : Frequency Domain Methods for Nonlinear Analysis : Theory and Applications, World Scientific Publ., Singapore, 1996; K.J. Astrom : Control of Complex Systems, Springer, Berlin, 2000; P.J. Antsaklis, K.M. Passino : An Introduction to Intelligent and Autonomous Control, Kluwer Acad. Publ., Norwell, MA, 1993; L.X. Wang : Adaptive Fuzzy Systems and Control : Design and Stability Analysis, Prentice−Hall, Englewood Cliffs, NJ, 1994; D. Yudin, A.Goryashko : Control Problems and Complexity Theory I, Engineering Cybernetics (USSR), Vol.12, pp.10-24, 1974, and II, Engineering Cybernetics (USSR), Vol.13, pp.1-13, 1975; ; K.A. Morris : Design of Finite-dimensional Controllers for Infinite-dimensional Systems by Approximation, J. Mathematical Systems, Estimation and Control, Vol.4(2), pp.1-30, 1994; M. Brown, C. Harris : Neuro Fuzzy Adaptive Modelling and Control, Prentice−Hall, Englewood Cliffs, NJ, 1994; B. Kosko : Neural Networks and Fuzzy Systems: A Dynamical Systems Approach to Machine Intelligence, Prentice−Hall, Englewood Cliffs, NJ, 1991; I. Grigorenko : Optimal Control and Forecasting of Complex Dynamical Systems, World Scientific, Singapore, 2006; C.I. Byrnes, C.F. Martin, R.E. Saeks (eds.) : Analysis and Control of Nonlinear Systems, North-Holland, Amsterdam, 1988; G. Conte, C.H. Moog, A.M. Perdon : Nonlinear Control Systems – An Algebraic Setting, Springer, London, 1998; G.W. Ng : Application of Neural Networks to Adaptive Control of Nonlinear Systems, Research Studies Press,

London, 1997; B.A. Francis : A Course in H∞ Control Theory, Springer−Verlag, Berlin, 1987; R.C. Dorf, R.H. Bishop : Modern Control Systems, Addison Wesley, Reading, Mass., 1995; J.W. Helton, M.R. James :

Extending H∞ Control to Nonlinear Systems, SIAM, Frontiers in Appl. Sci., 1999; V. Ionescu, A. Stoica : Robust Stabilization and H∞ Problems, Kluwer Acad. Publ., Dordrecht, The Netherlands, 1999; S.V. Emelyanov, A. Burovoi, F. Yu. Levada : Control of Indefinite Nonlinear Dynamic Systems : Induced Internal Feedback, Springer, London, 1998; G.E. Dullerud, F. Paganini : A Course in Robust Control Theory : A Convex Approach, Springer-Verlag, New−York, 2000; A.L.Fradkov, I.V. Miroshnik, V.O. Nikiforov : Nonlinear and Adaptive Control of Complex Systems, Kluwer Acad. Publ., Dordrecht, the Netherlands, 1999; M. Krstic, I. Kanellakopoulos, P.V. Kokotovic : Nonlinear and Adaptive Control Design, J. Wiley and Sons, New-York, 1995; B.A. Francis, A.R. Tanenbaum (eds.) : Feedback Control, Nonlinear Systems and Complexity, Springer, New-York, 1995; D. Schroeder (ed.) : Intelligent Observer and Control Design for Nonlinear Systems, Springer, New-York, 2000; V.V. Solodovnikov, V.I. Tumarkin : Complexity Theory and the Design of Control Systems, Theory and Methods of Systems Analysis, Vol.28, Nauka, Moscow, 1990; H.O. Wang, K. Tanaka, F. Griffin : An Approach to Fuzzy Control of Nonlinear Systems : Stability and Design Issues, IEEE Trans. On Fuzzy Systems, Vol.4(1), pp.14-23, 1996; M. Gopal : Modern Control System Theory, 2nd ed., J. Wiley and Sons, Toronto, Canada, 1993; L.X. Wang : Adaptive Fuzzy Systems and Control, Prentice Hall, Englewwod Cliffs, N.J., 1994 [28] K. Deimling : Nonlinear Functional Analysis, Springer−Verlag, Berlin, 1985; M.A. Krasnoselskii} : Asymptotics of Nonlinearities and Operator Equations, Birkhauser, Boston, Mass., 1995; J. Schwartz : Nonlinear Functional Analysis, Gordon and Breach, New−York, 1969; M.M. Vainberg : Variational Methods for the Study of Nonlinear Operators, Holden−Day, San Francisco, 1964; E. Zeidler : Nonlinear Functional Analysis and its Applications, Vol.I-III-IV, Springer−Verlag, New−York, 1985-86-88; J.S. Bendat : Nonlinear Systems Techniques and Applications, J. Wiley and Sons, New−York, 1998; N.A. Bobylev, Y.M. Burman, S.K.Korovin : Nonlinear Analysis and Applications, Walter de Gruyter, Berlin, 1994; V.N. Bogaesvski, A. Povzner : Algebraic Methods in Nonlinear Perturbation Theory, Springer−Verlag, New−York, 1991; L.A. Lusternik, V.I. Sobolev : Elements of Functional Analysis, Ungar, New−York, 1961; J.K. Hale, S.M. Verduyn Lunel : Theory of Functional Differential Equations, Springer−Verlag, New-York, 1993 [29] C. Corduneanu : Almost Periodic Functions, Interscience tracts in Pure and Applied Maths., Vol.22, Interscience, New−York, 1961; J.A. Dubinskij : Sobolev Spaces of Infinite Order and Differential Equations, Teubner, Leipzig, 1988; L. Amerio, G. Prouse : Almost−Periodic Functions and Functional Equations, Van Nostrand-Reinhold, New−York, 1971; V.G. Mazja : Sobolev Spaces, Springer−Verlag, New−York, 1985; M.M. Rao, Z.D. Ren : Theory of Orlicz Spaces, Marcel Dekker, New−York, 1991 [30] J. Dugundji, A. Granas : Fixed−Point Theory, Vol.I, Polish Scientific Publ., Warsaw, 1982; T.H. Jiang : Fixed−Point Theory, Springer−Verlag, New−York, 1985; E. Zeidler : Nonlinear Functional Analysis and its Applications, Vol.I, Springer−Verlag, New−York, 1986; D.R. Smart : Fixed Point Theorems, Cambridge Univ. Press, Mass., 1980; T.A. Burton : Stability by Fixed Point Theory for Functional Differential Equations, Dover Publ., New−York, 2006 [31] M. Cotsaftis : Recent Advances in Control of Complex Systems, Survey Lecture, Proc. ESDA'96, Montpellier, France, ASME, Vol.I, 1996, p.1; M. Cotsaftis : Comportement et Contrôle des Systèmes Complexes, Diderot, Paris, 1997; M. Cotsaftis : Popov Criterion Revisited for Other Nonlinear Systems, Proc. ISIC 2003 (International Symposium on Intelligent Control), Oct 5-8, Houston, 2003; [32] A.V. Renouzenko : Partial Differential Equations with Discrete and Distributed State−Dependent Delays, J. Math. Anal. Applic., Vol.326(2), pp.1031-1045, 2007 [33] J. Appell, P.P. Zabrijko : Nonlinear Superposition Operators, Cambridge University Press, Mass., 1990 [34] B.G. Pachpatte : Mathematical Inequalities, North Holland, Elsevier, Amsterdam, 2005; D. Bainov, P. Simeonov : Integral Inequalities and Applications, Kluwer Academic Press, Dordrecht, the Netherlands, 1992; Y.H. Kim : Explicit Bounds on Some Nonlinear Integral Inequalities, J. Inequalities in Pure and Applied Mathematics, Vol.7(2), 2006; D. S. Mitrinovic, J. E. Pecaric, A. M. Fink : Inequalities Involving Functions and Their Integrals and Derivatives, Kluwer Acad. Publ., Dordrecht, Boston, London, 1991 [35] M. Cotsaftis : Robust Asymptotically Stable Control for Intelligent Unknown Mechatronic Systems, Proc. ICARCV 2006, Singapore, Dec 5-8, 2006 [36] M. Cotsaftis : Beyond Mechatronics , Toward Global Machine Intelligence, Proc. ICMT’2005, Kuala– Lumpur, Dec. 6-8, 2005 [37] M. Cotsaftis : From Trajectory Control to Task Control – Emergence of Self Organization in Complex Systems, in Emergent Properties in Natural and Artificial Dynamical Systems, M.A. Aziz-Alaoui, C. Bertelle ( eds.), Springer−Verlag, pp.3-22, 2006; also : On the Definition of Task Oriented Intelligent Control, Proc. ISIC’02 Conf., Vancouver, Oct. 27-30, 2002 [38] M. Cotsaftis : Merging Information Technologies with Mechatronics – The Autonomous Intelligence Challenge, Proc. IEECON’2006, Nov.6-10, Paris, 2006